Home Explore Handbook of Practical Medicine Stroke

Handbook of Practical Medicine Stroke

Published by andiny.clock, 2014-07-25 10:34:00

Description: We don’t care much for abbreviations. They are not literate (Oliver Twist was not abbreviated to OT each time
Dickens mentioned his name!), they don’t look good on
the printed page, and they make things more difficult to
read and understand, particularly for non-experts. But
they do save space and so we have to use them a bit.
However, we will avoid them as far as we can in tables,
figures and the practice points. We will try to define any
abbreviations the first time they are used in each chapter, or even in each section if they are not very familiar.
But, if we fail to be comprehensible, then here is a rather
long list to refer to.

Read the Text Version

Pages:

9781405127660_4_009.qxd 10/13/07 11:03 AM Page 493 References 493 ruptured vertebrobasilar dissecting aneurysms. 115 Al-Shahi R, Bhattacharya JJ, Currie DG, Papanastassiou V, Neurosurgery 1995; 36:905–13. Ritchie V, Roberts RC et al. Prospective, population-based 99 Kawamata T, Tanikawa T, Takeshita M, Onda H, detection of intracranial vascular malformations in adults: Takakura K, Toyoda C. Rebleeding of intracranial the Scottish Intracranial Vascular Malformation Study dissecting aneurysm in the vertebral artery following (SIVMS). Stroke 2003; 34:1163–9. proximal clipping. Neurol Res 1994; 16(2):141–4. 116 Stapf C, Mast H, Sciacca RR, Berenstein A, Nelson PK, 100 Nakatomi H, Nagata K, Kawamoto S, Shiokawa Y. Ruptured Gobin YP et al. The New York Islands AVM Study: design, dissecting aneurysm as a cause of subarachnoid study progress, and initial results. Stroke 2003; hemorrhage of unveriﬁed etiology. Stroke 1997; 34:e29–e33. 28:1278–82. 117 Linn FHH, Rinkel GJE, Algra A, van Gijn J. Incidence of 101 Ohkuma H, Suzuki S, Ogane K. Dissecting aneurysms of subarachnoid hemorrhage: Role of region, year, and rate intracranial carotid circulation. Stroke 2002; 33:941–7. of computed tomography: a meta-analysis. Stroke 1996; 102 Adams HP, Jr., Aschenbrener CA, Kassell NF, Ansbacher L, 27:625–9. Cornell SH. Intracranial hemorrhage produced by 118 Van Gijn J, van Dongen KJ. The time course of aneurysmal spontaneous dissecting intracranial aneurysm. Arch Neurol haemorrhage on computed tomograms. Neuroradiology 1982; 39(12):773–6. 1982; 23(3):153–6. 103 Massoud TF, Anslow P, Molyneux AJ. Subarachnoid 119 Abbed KM, Ogilvy CS. Intracerebral hematoma from hemorrhage following spontaneous intracranial carotid aneurysm rupture. Neurosurg Focus 2003; 15(4):1–5. artery dissection. Neuroradiology 1992; 34(1):33–5. 120 Kloster R. [Subarachnoid hemorrhage in Vestfold county. 104 Ohkuma H, Nakano T, Manabe H, Suzuki S. Subarachnoid Occurrence and prognosis]. Tidsskr Nor Laegeforen 1997; hemorrhage caused by a dissecting aneurysm of the 117:1879–82. internal carotid artery. J Neurosurg 2002; 97:576–83. 121 Brown RD, Jr., Wiebers DO, Forbes GS. Unruptured 105 Kunze S, Schiefer W. Angiographic demonstration of a intracranial aneurysms and arteriovenous malformations: dissecting aneurysm of the middle cerebral artery. frequency of intracranial hemorrhage and relationship of Neuroradiology 1971; 2(4):201–6. lesions. J Neurosurg 1990; 73(6):859–63. 106 Sasaki O, Koike T, Tanaka R, Ogawa H. Subarachnoid 122 Marks MP, Lane B, Steinberg GK, Snipes GJ. Intranidal hemorrhage from a dissecting aneurysm of the middle aneurysms in cerebral arteriovenous malformations: cerebral artery: case report. J Neurosurg 1991; 74(3):504–7. evaluation and endovascular treatment. Radiology 1992; 107 Piepgras DG, McGrail KM, Tazelaar HD. Intracranial 183(2):355–60. dissection of the distal middle cerebral artery as an 123 Stapf C, Mohr JP, Pile-Spellman J, Sciacca RR, Hartmann A, uncommon cause of distal cerebral artery aneurysm: Schumacher HC et al. Concurrent arterial aneurysms in case report. J Neurosurg 1994; 80(5):909–13. brain arteriovenous malformations with haemorrhagic 108 Ohkuma H, Suzuki S, Shimamura N, Nakano T. presentation. J Neurol Neurosurg Psychiatry 2002; 73:294–8. Dissecting aneurysms of the middle cerebral artery: 124 Kim EJ, Halim AX, Dowd CF, Lawton MT, Singh V, Bennett neuroradiological and clinical features. Neuroradiology J et al. The relationship of coexisting extranidal aneurysms 2003; 45:143–8. to intracranial hemorrhage in patients harboring brain 109 Guridi J, Gallego J, Monzon F, Aguilera F. Intracerebral arteriovenous malformations. Neurosurgery 2004; hemorrhage caused by transmural dissection of the 54:1349–57. anterior cerebral artery. Stroke 1993; 24(9):1400–2. 125 Nakatomi H, Segawa H, Kurata A, Shiokawa Y, Nagata K, 110 Koyama S, Kotani A, Sasaki J. Spontaneous dissecting Kamiyama H et al. Clinicopathological study of aneurysm of the anterior cerebral artery: report of two intracranial fusiform and dolichoectatic aneurysms: cases. Surg Neurol 1996; 46:55–61. insight on the mechanism of growth. Stroke 2000; 111 Mori K, Yamamoto T, Maeda M. Dissecting aneurysm 31:896–900. conﬁned to the anterior cerebral artery. Br J Neurosurg 126 Passero SG, Calchetti B, Bartalini S. Intracranial bleeding in 2002; 16:158–64. patients with vertebrobasilar dolichoectasia. Stroke 2005; 112 Ohkuma H, Suzuki S, Kikkawa T, Shimamura N. 36:1421–5. Neuroradiologic and clinical features of arterial dissection 127 Ricolﬁ F, Decq P, Brugieres P, Blustajn J, Melon E, Gaston of the anterior cerebral artery. Am J Neuroradiol 2003; A. Ruptured fusiform aneurysm of the superior third of the 24:691–9. basilar artery associated with the absence of the midbasilar 113 Kurino M, Yoshioka S, Ushio Y. Spontaneous dissecting artery: case report. J Neurosurg 1996; 85:961–5. aneurysms of anterior and middle cerebral artery 128 Flemming KD, Josephs K, Wijdicks EFM. Enlarging associated with brain infarction: a case report and review vertebrobasilar dolichoectasia with subarachnoid of the literature. Surg Neurol 2002; 57:428–36. hemorrhage heralded by recurrent ischemia: case 114 Aoki N. Do intracranial arteriovenous malformations cause illustration. J Neurosurg 2000; 92:504. subarachnoid haemorrhage? Review of computed 129 Flemming KD, Wiebers DO, Brown RD, Jr., Link MJ, tomography features of ruptured arteriovenous Nakatomi H, Huston J, III et al. Prospective risk of malformations in the acute stage. Acta Neurochir (Wien) hemorrhage in patients with vertebrobasilar nonsaccular 1991; 112(3–4):92–5. intracranial aneurysm. J Neurosurg 2004; 101:82–7. .. ..

9781405127660_4_009.qxd 10/13/07 11:03 AM Page 494 494 Chapter 9 What caused this subarachnoid haemorrhage? 130 Day AL, Gaposchkin CG, Yu CJ, Rivet DJ, Dacey RG, Jr. 145 Chang R, Friedman DP. Isolated cortical venous Spontaneous fusiform middle cerebral artery aneurysms: thrombosis presenting as subarachnoid hemorrhage: characteristics and a proposed mechanism of formation. a report of three cases. Am J Neuroradiol 2004; J Neurosurg 2003; 99:228–40. 25:1676–9. 131 Horie N, Takahashi N, Furuichi S, Mori K, Onizuka M, 146 Oppenheim C, Domigo V, Gauvrit JY, Lamy C, kowiak- Tsutsumi K et al. Giant fusiform aneurysms in the middle Cordoliani MA, Pruvo JP et al. Subarachnoid hemorrhage cerebral artery presenting with hemorrhages of different as the initial presentation of dural sinus thrombosis. origins: report of three cases and review of the literature. Am J Neuroradiol 2005; 26:614–17. J Neurosurg 2003; 99:391–6. 147 Nakahara K, Oka H, Utsuki S, Lida H, Kurita M, Mochizuki 132 Fuentes S, Levrier O, Metellus P, Dufour H, Fuentes JM, T et al. Pituitary apoplexy manifesting as diffuse Grisoli F. Giant fusiform intracranial A2 aneurysm: subarachnoid hemorrhage. Neurol Med Chir (Tokyo) 2006; endovascular and surgical treatment. Case illustration. 46:594–7. J Neurosurg 2004; 101:704. 148 Satyarthee GD, Mahapatra AK. Pituitary apoplexy in a 133 Selviaridis P, Spiliotopoulos A, Antoniadis C, Kontopoulos child presenting with massive subarachnoid and V, Foroglou G. Fusiform aneurysm of the posterior cerebral intraventricular hemorrhage. Clin Neurosci 2005; 12:94–6. artery: report of two cases. Acta Neurochir (Wien) 2002; 149 Wohaibi MA, Russell NA, Ferayan AA, Awada A, Jumah 144:295–9. MA, Omojola M. Pituitary apoplexy presenting as massive 134 Zicherman J, Roychowdhury S, Demarco JK, Shepard S, subarachnoid hemorrhage. J Neurol Neurosurg Psychiatry Schonfeld S, Keller I et al. Endovascular treatment of a 2000; 69:700–1. ruptured giant serpentine aneurysm of the superior 150 Biousse V, Newman NJ, Oyesiku NM. Precipitating factors cerebellar artery in a patient with a Chiari II malformation. in pituitary apoplexy. J Neurol Neurosurg Psychiatry 2001; Am J Neuroradiol 2004; 25:1077–9. 71:542–5. 135 Sarkar A, Link MJ. Distal anterior inferior cerebellar artery 151 Reid RL, Quigley ME, Yen SS. Pituitary apoplexy: a review. aneurysm masquerading as a cerebellopontine angle Arch Neurol 1985; 42(7):712–19. tumor: case report and review of literature. Skull Base 152 Dodick DW, Wijdicks EFM. Pituitary apoplexy presenting 2004; 14:101–6. as a thunderclap headache. Neurology 1998; 50:1510–11. 136 Liew D, Ng PY, Ng I. Surgical management of ruptured and 153 Al Wohaibi M, Russell NA, Al Ferayan A, Awada A, Al unruptured symptomatic posterior inferior cerebellar Jumah M, Omojola M. Pituitary apoplexy presenting as artery aneurysms. Br J Neurosurg 2004; 18:608–12. massive subarachnoid haemorrhage. J Neurol Neurosurg 137 Josephson SA, Johnston SC. Multiple stable fusiform Psychiatry 2000; 69:700–1. intracranial aneurysms following atrial myxoma. Neurology 154 Noordzij MJ, de Heide LJ, van den BG, Hoving EW. 2005; 64:526. [Pituitary apoplexy: an endocrinologic emergency] 138 Lasjaunias P, Chiu M, ter Brugge K, Tolia A, Hurth M, Apoplexie van de hypofyse: een endocrinologisch Bernstein M. Neurological manifestations of intracranial spoedgeval. Ned Tijdschr Geneeskd 2005; 149:2748–51. dural arteriovenous malformations. J Neurosurg 1986; 155 McFadzean RM, Doyle D, Rampling R, Teasdale E, Teasdale 64(5):724–30. G. Pituitary apoplexy and its effect on vision. Neurosurgery 139 Duffau H, Lopes M, Janosevic V, Sichez JP, Faillot T, 1991; 29(5):669–75. Capelle L et al. Early rebleeding from intracranial dural 156 Gazioglu N, Kadioglu P, Ocal E, Erman H, Akar Z, Oz B. An arteriovenous ﬁstulas: report of 20 cases and review of the unusual presentation of Nelson’s syndrome with apoplexy literature. J Neurosurg 1999; 90:78–84. and subarachnoid hemorrhage. Pituitary 2002; 5:267–74. 140 Halbach VV, Higashida RT, Hieshima GB, Goto K, 157 Arita K, Kurisu K, Tominaga A, Sugiyama K, Ikawa F, Norman D, Newton TH. Dural ﬁstulas involving the Yoshioka H et al. Thickening of sphenoid sinus mucosa transverse and sigmoid sinuses: results of treatment in during the acute stage of pituitary apoplexy: case report. 28 patients. Radiology 1987; 163(2):443–7. J Neurosurg 2001; 95:897–901. 141 Ciccone A, Citterio A, Santilli I, Sterzi R. Subarachnoid 158 Brust JC, Dickinson PC, Hughes JE, Holtzman RN. The haemorrhage treated with anticoagulants. Lancet 2000; diagnosis and treatment of cerebral mycotic aneurysms. 356:1818. Ann Neurol 1990; 27(3):238–46. 142 De Bruijn SFTM, Stam J, Kappelle LJ. Thunderclap 159 Masuda J, Yutani C, Waki R, Ogata J, Kuriyama Y, headache as ﬁrst symptom of cerebral venous sinus Yamaguchi T. Histopathological analysis of the thrombosis. Lancet 1996; 348:1623–5. mechanisms of intracranial hemorrhage complicating 143 Carviy Nievas M, Haas E, Hollerhage HG, Lorey T, Klein PJ. infective endocarditis. Stroke 1992; 23(6):843–50. Cerebral vein thrombosis associated with aneurysmal 160 Chukwudelunzu FE, Brown RD, Jr., Wijdicks EF, subarachnoid bleeding: implications for treatment. Surg Steckelberg JM. Subarachnoid haemorrhage associated Neurol 2004; 61:95–8. with infectious endocarditis: case report and literature 144 Adaletli I, Sirikci A, Kara B, Kurugoglu S, Ozer H, Bayram review. Eur J Neurol 2002; 9:423–7. M. Cerebral venous sinus thrombosis presenting with 161 Krapf H, Skalej M, Voigt K. Subarachnoid hemorrhage due excessive subarachnoid hemorrhage in a 14-year-old boy. to septic embolic infarction in infective endocarditis. Emerg Radiol 2005; 12:57–9. Cerebrovasc Dis 1999; 9:182–4. .. ..

9781405127660_4_009.qxd 10/13/07 11:03 AM Page 495 References 495 162 Kowall NW, Sobel RA. Case records of the Massachusetts 180 Levine SR, Brust JC, Futrell N, Ho KL, Blake D, Millikan CH General Hospital. Weekly clinicopathological exercises. et al. Cerebrovascular complications of the use of the Case 7-1988. A 27-year-old man with acute ‘crack’ form of alkaloidal cocaine. N Engl J Med 1990; myelomonocytic leukemia in remission and repeated 323(11):699–704. intracranial hemorrhages. N Engl J Med 1988; 181 Nanda A, Vannemreddy PSSV, Polin RS, Willis BK. 318(7):427–40. Intracranial aneurysms and cocaine abuse: analysis of 163 Lau AHC, Takeshita M, Ishii M. Mycotic (Aspergillus) prognostic indicators. Neurosurgery 2000; 46:1063–7. arteriitis resulting in fatal subarachnoid hemorrhage: 182 Simpson RK, Jr., Fischer DK, Narayan RK, Cech DA, a case report. Angiology 1991; 42:251–5. Robertson CS. Intravenous cocaine abuse and 164 Ho CL, Deruytter MJ. CNS aspergillosis with mycotic subarachnoid haemorrhage: effect on outcome. aneurysm, cerebral granuloma and infarction. Acta Br J Neurosurg 1990; 4(1):27–30. Neurochir (Wien) 2004; 146:851–6. 183 Oyesiku NM, Colohan AR, Barrow DL, Reisner A. Cocaine- 165 Young RC, Bennett JE, Vogel CL, Carbone PP, DeVita VT. induced aneurysmal rupture: an emergent negative factor Aspergillosis. The spectrum of the disease in 98 patients. in the natural history of intracranial aneurysms? Medicine (Baltimore) 1970; 49(2):147–73. Neurosurgery 1993; 32:518–25. 166 Hope WW, Walsh TJ, Denning DW. Laboratory diagnosis 184 Krendel DA, Ditter SM, Frankel MR, Ross WK. Biopsy- of invasive aspergillosis. Lancet Infect Dis 2005; 5:609–22. proven cerebral vasculitis associated with cocaine abuse. 167 Knouse MC, Madeira RG, Celani VJ. Pseudomonas Neurology 1990; 40(7):1092–4. aeruginosa causing a right carotid artery mycotic aneurysm 185 Boco T, Macdonald RL. Absence of acute cerebral after a dental extraction procedure. Mayo Clin Proc 2002; vasoconstriction after cocaine-associated subarachnoid 77:1125–30. hemorrhage. Neurocrit Care 2004; 1:449–54. 168 Furuya K, Sasaki T, Yoshimoto Y, Okada Y, Fujimaki T, 186 Bruno A, Nolte KB, Chapin J. Stroke associated with Kirino T. Histologically veriﬁed cerebral aneurysm ephedrine use. Neurology 1993; 43(7):1313–16. formation secondary to embolism from cardiac myxoma: 187 Auer J, Berent R, Weber T, Lassnig E, Eber B. Subarachnoid case report. J Neurosurg 1995; 83:170–3. haemorrhage with ‘ecstasy’ abuse in a young adult. Neurol 169 Bobo H, Evans OB. Intracranial aneurysms in a child with Sci 2002; 23:199–201. recurrent atrial myxoma. Pediatr Neurol 1987; 3:230–2. 188 Rinkel GJE, Prins NEM, Algra A. Outcome of aneurysmal 170 Van Beijnum J, Straver DCG, Rinkel GJE, Klijn CJM. subarachnoid hemorrhage in patients on anticoagulant Spinal arteriovenous shunts presenting as intracranial treatment. Stroke 1997; 28:6–9. subarachnoid haemorrhage. J Neurol 2007; [Epub ahead of 189 Mattle H, Kohler S, Huber P, Rohner M, Steinsiepe KF. print]. Anticoagulation-related intracranial extracerebral 171 Aminoff MJ, Logue V. Clinical features of spinal vascular haemorrhage. J Neurol Neurosurg Psychiatry 1989; malformations. Brain 1974; 97(1):197–210. 52(7):829–37. 172 Rengachary SS, Duke DA, Tsai FY, Kragel PJ. Spinal arterial 190 Papa ML, Schisano G, Franco A, Nina P. Congenital aneurysm: case report. Neurosurgery 1993; 33:125–9. deﬁciency of factor VII in subarachnoid hemorrhage. 173 Gonzalez LF, Zabramski JM, Tabrizi P, Wallace RC, Stroke 1994; 25(2):508–10. Massand MG, Spetzler RF. Spontaneous spinal 191 Cummings TJ, Johnson RR, Diaz FG, Michael DB. The subarachnoid hemorrhage secondary to spinal aneurysms: relationship of blunt head trauma, subarachnoid diagnosis and treatment paradigm. Neurosurgery 2005; hemorrhage, and rupture of pre-existing intracranial 57:1127–31. saccular aneurysms. Neurol Res 2000; 22:165–70. 174 Sakamoto M, Watanabe T, Okamoto H. A case of ruptured 192 Sakas DE, Dias LS, Beale D. Subarachnoid haemorrhage aneurysm associated with spinal arteriovenous presenting as head injury. Br Med J 1995; 310:1186–7. malformation presenting with hematomyelia: case report. 193 Sahjpaul RL, Abdulhak MM, Drake CG, Hammond RR. Surg Neurol 2002; 57:438–42. Fatal traumatic vertebral artery aneurysm rupture: case 175 Ohmori Y, Hamada J, Morioka M, Yoshida A. Spinal report. J Neurosurg 1998; 89:822–4. aneurysm arising from the feeding pedicle of a thoracic 194 Rosenow J, Das K, Weitzner I, Couldwell WT. Rupture of a perimedullary arteriovenous ﬁstula: case report. Surg Neurol large ophthalmic segment saccular aneurysm associated 2005; 64:468–70. with closed head injury: case report. Neurosurgery 2000; 176 Anson JA, Koshy M, Ferguson L, Crowell RM. 46:1515–17. Subarachnoid hemorrhage in sickle-cell disease. J Neurosurg 195 Tomlinson BE, Walton JN. Superﬁcial haemosiderosis of 1991; 75(4):552–8. the central nervous system. J Neurol Neurosurg Psychiatry 177 Preul M, Cendes F, Just N, Mohr G. Intracranial aneurysms 1964; 27:332–9. and sickle cell anemia: multiplicity and propensity for the 196 Fearnley JM, Stevens JM, Rudge P. Superﬁcial siderosis vertebrobasilar territory. Neurosurgery 1998; 42:971–7. of the central nervous system. Brain 1995; 178 Carey J, Numaguchi Y, Nadell J. Subarachnoid hemorrhage 118(4):1051–66. in sickle cell disease. Childs Nerv Syst 1990; 6(1):47–50. 197 Van Harskamp NJ, Rudge P, Cipolotti L. Cognitive and 179 Pegelow CH. Stroke in children with sickle cell anaemia: social impairments in patients with superﬁcial siderosis. aetiology and treatment. Paediatr Drugs 2001; 3:421–32. Brain 2005; 128:1082–92. .. ..

9781405127660_4_009.qxd 10/13/07 11:03 AM Page 496 496 Chapter 9 What caused this subarachnoid haemorrhage? 198 Kumar N, Lindell EP, Wilden JA, Davis DH. Role of correlations. Comput Med Imaging Graph 1991; dynamic CT myelography in identifying the etiology 15(1):43–51. of superﬁcial siderosis. Neurology 2005; 65:486–8. 215 Young WB, Lee KP, Pessin MS, Kwan ES, Rand WM, 199 Bonito V, Agostinis C, Ferraresi S, Defanti CA. Superﬁcial Caplan LR. Prognostic signiﬁcance of ventricular blood in siderosis of the central nervous system after brachial plexus supratentorial hemorrhage: a volumetric study. Neurology injury: case report. J Neurosurg 1994; 80(5):931–4. 1990; 40(4):616–19. 200 Konitsiotis S, Argyropoulou MI, Kosta P, Giannopoulou M, 216 Roos YB, Hasan D, Vermeulen M. Outcome in patients Efremidis SC, Kyritsis AP. CNS siderosis after brachial with large intraventricular haemorrhages: a volumetric plexus avulsion. Neurology 2002; 58:505. study. J Neurol Neurosurg Psychiatry 1995; 58:622–4. 201 Li KW, Haroun RI, Clatterbuck RE, Murphy K, Rigamonti 217 Verma A, Maheshwari MC, Bhargava S. Spontaneous D. Superﬁcial siderosis associated with multiple cavernous intraventricular haemorrhage. J Neurol 1987; 234(4):233–6. malformations: report of three cases. Neurosurgery 2001; 218 Waga S, Shimosaka S, Kojima T. Arteriovenous 48:1147–50. malformations of the lateral ventricle. J Neurosurg 1985; 202 Padberg M, Hoogenraad TU. Cerebral siderosis: deafness 63(2):185–92. by a spinal tumour. J Neurol 2000; 247:473. 219 Darby DG, Donnan GA, Saling MA, Walsh KW, Bladin PF. 203 Anderson NE, Shefﬁeld S, Hope JKA. Superﬁcial siderosis Primary intraventricular hemorrhage: clinical and of the central nervous system: a late complication of neuropsychological ﬁndings in a prospective stroke series. cerebellar tumors. Neurology 1999; 52:163–9. Neurology 1988; 38(1):68–75. 204 McCarron MO, Flynn PA, Owens C, Wallace I, Mirakhur 220 Donnet A, Balzamo M, Royere ML, Grisoli F, Ali Cherif A. M, Gibson JM et al. Superﬁcial siderosis of the central [Transient Korsakoff’s syndrome after intraventricular nervous system many years after neurosurgical procedures. hemorrhage] Syndrome de Korsakoff transitoire au décours J Neurol Neurosurg Psychiatry 2003; 74:1326–8. d’une hémorragie intraventriculaire. Neurochirurgie 1992; 205 Mascalchi M, Salvi FP, Pirini MG, D’Errico A, Ferlini A, 38(2):102–4. Lolli F et al. Transthyretin amyloidosis and superﬁcial 221 Caesar B, Middleton PM, Watkins LD. Sudden onset siderosis of the CNS. Neurology 1999; 53:1498–503. of vomiting as a presentation of perimesencephalic 206 Jin K, Sato S, Takahashi T, Nakazaki H, Date Y, subarachnoid haemorrhage. Eur J Emerg Med 2005; Nakazato M et al. Familial leptomeningeal amyloidosis 12:185–7. with a transthyretin variant Asp18Gly representing 222 Hart RG, Byer JA, Slaughter JR, Hewett JE, Easton JD. repeated subarachnoid haemorrhages with superﬁcial Occurrence and implications of seizures in subarachnoid siderosis. J Neurol Neurosurg Psychiatry 2004; 75:1463–6. hemorrhage due to ruptured intracranial aneurysms. 207 Uchino A, Aibe H, Itoh H, Aiko Y, Tanaka M. Superﬁcial Neurosurgery 1981; 8(4):417–21. siderosis of the central nervous system: its MRI 223 Pinto AN, Canhao P, Ferro JM. Seizures at the onset of manifestations. Clin Imaging 1997; 21:241–5. subarachnoid haemorrhage. J Neurol 1996; 208 Canhao P, Ferro JM, Pinto AM, Melo TP, Campos JG. 243(2):161–64. Perimesencephalic and nonperimesencephalic 224 Claassen J, Peery S, Kreiter KT, Hirsch LJ, Du EY, Connolly subarachnoid haemorrhages with negative angiograms. ES et al. Predictors and clinical impact of epilepsy after Acta Neurochir (Wien) 1995; 132:14–19. subarachnoid hemorrhage. Neurology 2003; 60:208–14. 209 Ruigrok YM, Rinkel GJE, van Gijn J. CT patterns and 225 Polmear A. Sentinel headaches in aneurysmal long-term outcome in patients with an aneurysmal subarachnoid haemorrhage: what is the true incidence? type of subarachnoid hemorrhage and repeatedly A systematic review. Cephalalgia 2003; 23:935–41. negative angiograms. Cerebrovasc Dis 2002; 14:221–7. 226 Linn FHH, Wijdicks EFM, van der Graaf Y, Weerdesteyn- 210 Du Mesnil de Rochemont R, Heindel W, Wesselmann C, van Vliet FA, Bartelds AI, van Gijn J. Prospective study Krüger K, Lanfermann H, Ernestus RI et al. Nontraumatic of sentinel headache in aneurysmal subarachnoid subarachnoid hemorrhage: value of repeat angiography. haemorrhage. Lancet 1994; 344(8922):590–3. Radiology 1997; 202(3):798–800. 227 Linn FHH, Rinkel GJE, Algra A, van Gijn J. The notion of 211 Moritake K, Handa H, Ohtsuka S, Hashimoto N. Vanishing ‘warning leaks’ in subarachnoid haemorrhage: are such cerebral aneurysm in serial angiography. Surg Neurol 1981; patients in fact admitted with a rebleed? J Neurol Neurosurg 16(1):36–40. Psychiatry 2000; 68:332–6. 212 Kaim A, Proske M, Kirsch E, von Weymarn A, Radu EW, 228 Chaudhary MY, Sachdev VP, Cho SH, Weitzner I, Jr., Steinbrich W. Value of repeat-angiography in cases of Puljic S, Huang YP. Dural arteriovenous malformation unexplained subarachnoid hemorrhage (SAH). of the major venous sinuses: an acquired lesion. Am J Acta Neurol Scand 1996; 93(5):366–73. Neuroradiol 1982; 3(1):13–19. 213 Iwanaga H, Wakai S, Ochiai C, Narita J, Inoh S, Nagai M. 229 Salgado AV, Furlan AJ, Keys TF. Mycotic aneurysm, Ruptured cerebral aneurysms missed by initial subarachnoid hemorrhage, and indications for cerebral angiographic study. Neurosurgery 1990; angiography in infective endocarditis. Stroke 1987; 27(1):45–51. 18(6):1057–60. 214 Weisberg LA, Elliott D, Shamsnia M. Intraventricular 230 Reijneveld JC, Wermer MJH, Boonman Z, van Gijn J, hemorrhage in adults: clinical-computed tomographic Rinkel GJE. Acute confusional state as presenting feature .. ..

9781405127660_4_009.qxd 10/13/07 11:03 AM Page 497 References 497 in aneurysmal subarachnoid hemorrhage: frequency and 247 Date I, Akioka T, Ohmoto T. Penetration of the optic characteristics. J Neurol 2000; 247:112–16. chiasm by a ruptured anterior communicating artery 231 Caeiro L, Menger C, Ferro JM, Albuquerque R, Figueira ML. aneurysm: case report. J Neurosurg 1997; 87:324–6. Delirium in acute subarachnoid haemorrhage. Cerebrovasc 248 Hyland HH, Barnett HJM. The pathogenesis of cranial Dis 2005; 19:31–8. nerve palsies associated with intracranial aneurysms. 232 Benbadis SR, Sila CA, Cristea RL. Mental status changes Proc Roy Soc Med 1954; 47:141–6. and stroke. J Gen Intern Med 1994; 9:485–7. 249 Vincent FM, Zimmerman JE. Superior cerebellar artery 233 Pfausler B, Belcl R, Metzler R, Mohsenipour I, Schmutzhard aneurysm presenting as an oculomotor nerve palsy in a E. Terson’s syndrome in spontaneous subarachnoid child. Neurosurgery 1980; 6(6):661–4. hemorrhage: a prospective study in 60 consecutive 250 Kamat AA, Tizzard S, Mathew B. Painful third nerve palsy patients. J Neurosurg 1996; 85:392–4. in a patient with perimesencephalic subarachnoid 234 Frizzell RT, Kuhn F, Morris R, Quinn C, Fisher WS, III. haemorrhage. Br J Neurosurg 2005; 19:247–50. Screening for ocular hemorrhages in patients with 251 Nadeau SE, Trobe JD. Pupil sparing in oculomotor palsy: ruptured cerebral aneurysms: a prospective study of 99 a brief review. Ann Neurol 1983; 13(2):143–8. patients. Neurosurgery 1997; 41:529–33. 252 Kissel JT, Burde RM, Klingele TG, Zeiger HE. Pupil-sparing 235 Stiebel-Kalish H, Turtel LS, Kupersmith MJ. The natural oculomotor palsies with internal carotid-posterior history of nontraumatic subarachnoid hemorrhage-related communicating artery aneurysms. Ann Neurol 1983; intraocular hemorrhages. Retina 2004; 24:36–40. 13(2):149–54. 236 Kiriakopoulos ET, Gorn RA, Barton JJ. Small retinal 253 Keane JR. Aneurysmal third-nerve palsies presenting with hemorrhages as the only sign of an intracranial aneurysm. pleocytosis. Neurology 1996; 46:1176. Am J Ophthalmol 1998; 125:401–3. 254 Van Gijn J, Hijdra A, Wijdicks EFM, Vermeulen M, 237 Murthy S, Salas D, Hirekataur S, Ram R. Terson’s syndrome van Crevel H. Acute hydrocephalus after aneurysmal presenting as an ophthalmic emergency. Acta Ophthalmol subarachnoid hemorrhage. J Neurosurg 1985; 63(3):355–62. Scand 2002; 80:665–6. 255 Maramattom BV, Wijdicks EFM. Dorsal mesencephalic 238 Manconi M, Paolino E, Casetta I, Granieri E. Anosmia in syndrome and acute hydrocephalus after subarachnoid a giant anterior communicating artery aneurysm. Arch hemorrhage. Neurocrit Care 2005; 3:57–8. Neurol 2001; 58:1474–5. 256 Swash M. Periaqueductal dysfunction (the Sylvian 239 Haritoglou C, Müller-Schunk S, Weber C, Hoffmann U, aqueduct syndrome): a sign of hydrocephalus? J Neurol Ulbig MW. Central retinal artery occlusion in association Neurosurg Psychiatry 1974; 37(1):21–6. with an aneurysm of the internal carotid artery. Am J 257 Fisher CM. Clinical syndromes in cerebral thrombosis, Ophthalmol 2001; 132:270–1. hypertensive hemorrhage, and ruptured saccular 240 Kurita H, Kawamoto S, Ueki K, Kirino T. Dejerine aneurysm. Clin Neurosurg 1975; 22:117–47. syndrome caused by an aneurysmal compression. Arch 258 Senter HJ, Sarwar M. Nontraumatic dissecting aneurysm Neurol 2000; 57:1639–40. of the vertebral artery. J Neurosurg 1982; 56:128–30. 241 Mizobuchi M, Ito N, Tanaka C, Sako K, Sumi Y, Sasaki T. 259 Sarner M, Rose FC. Clinical presentation of ruptured Unidirectional olfactory hallucination associated with intracranial aneurysm. J Neurol Neurosurg Psychiatry 1967; ipsilateral unruptured intracranial aneurysm. Epilepsia 30:67–70. 1999; 40:516–19. 260 Endo H, Shimizu H, Tominaga T. Paraparesis associated 242 Friedman JA, Piepgras DG, Pichelmann MA, Hansen KK, with ruptured anterior cerebral artery territory aneurysms. Brown RD, Jr., Wiebers DO. Small cerebral aneurysms Surg Neurol 2005; 64:135–9. presenting with symptoms other than rupture. Neurology 261 Greene KA, Marciano FF, Dickman CA, Coons SW, 2001; 57:1212–16. Johnson PC, Bailes JE et al. Anterior communicating artery 243 Yanaka K, Matsumaru Y, Mashiko R, Hyodo A, Sugimoto K, aneurysm paraparesis syndrome: clinical manifestations Nose T. Small unruptured cerebral aneurysms presenting and pathologic correlates. Neurology 1995; 45(1):45–50. with oculomotor nerve palsy. Neurosurgery 2003; 262 Van de Warrenburg B, Vos P, Merx H, Kremer B. 52:553–6. Paroxysmal leg weakness and hearing loss in a patient with 244 Tummala RP, Harrison A, Madison MT, Nussbaum ES. subarachnoid hemorrhage. Eur Neurol 2000; 44:186–7. Pseudomyasthenia resulting from a posterior carotid artery 263 Ferrante L, Acqui M, Mastronardi L, Celli P, Lunardi P, wall aneurysm: a novel presentation: case report. Fortuna A. Posterior inferior cerebellar artery (PICA) Neurosurgery 2001; 49:1466–8. aneurysm presenting with SAH and contralateral crural 245 Chan JW, Hoyt WF, Ellis WG, Gress D. Pathogenesis of monoparesis: a case report. Surg Neurol 1992; 38:43–45. acute monocular blindness from leaking anterior 264 Laissy JP, Normand G, Monroc M, Duchateau C, Alibert F, communicating artery aneurysms: report of six cases. Thiebot J. Spontaneous intracerebral hematomas from Neurology 1997; 48(3):680–3. vascular causes. Predictive value of CT compared with 246 Hara N, Mukuno K, Ohtaka H, Shimizu K. Ischemic optic angiography. Neuroradiology 1991; 33(4):291–5. neuropathy associated with subarachnoid hemorrhage 265 Hillman J. Selective angiography for early aneurysm after rupture of anterior communicating artery aneurysm. detection in acute subarachnoid haemorrhage. Acta Ophthalmologica 2003; 217:79–84. Neurochir (Wien) 1993; 121:20–5. .. ..

9781405127660_4_009.qxd 10/13/07 11:03 AM Page 498 498 Chapter 9 What caused this subarachnoid haemorrhage? 266 Kutluhan S, Oyar O, Ozmen S, Noyaner A, Guler K, 282 Kershenovich A, Rappaport ZH, Maimon S. Brain Yesildag A. The unusually shaped bifrontal hematoma. computed tomography angiographic scans as the sole Stroke 2002; 33:876–7. diagnostic examination for excluding aneurysms in 267 Jackson A, Fitzgerald JB, Hartley RW, Leonard A, Yates J. patients with perimesencephalic subarachnoid CT appearances of haematomas in the corpus callosum in hemorrhage. Neurosurgery 2006; 59:798–801. patients with subarachnoid haemorrhage. Neuroradiology 283 Ruigrok YM, Rinkel GJE, Buskens E, Velthuis BK, van Gijn 1993; 35(6):420–3. J. Perimesencephalic hemorrhage and CT angiography: 268 Van der Jagt M, Hasan D, Bijvoet HWC, Pieterman H, a decision analysis. Stroke 2000; 31:2976–83. Dippel DWJ, Vermeij FH et al. Validity of prediction of the 284 Velthuis BK, Rinkel GJE, Ramos LMP, Witkamp TD, Van site of ruptured intracranial aneurysms with CT. Neurology Leeuwen MS. Perimesencephalic hemorrhage: exclusion 1999; 52:34–9. of vertebrobasilar aneurysms with CT angiography. Stroke 269 Hino A, Fujimoto M, Iwamoto Y, Yamaki T, Katsumori T. 1999; 30:1103–9. False localization of rupture site in patients with multiple 285 Lavin PJ, Troost BT. Traumatic fourth nerve palsy. cerebral aneurysms and subarachnoid hemorrhage. Clinicoanatomic correlations with computed Neurosurgery 2000; 46:825–30. tomographic scan. Arch Neurol 1984; 41(6):679–80. 270 Barker RA, Phillips RR, Moseley IF, Taylor WJ, Kitchen ND, 286 Takenaka N, Mine T, Suga S, Tamura K, Sagou M, Hirose Y Scadding JW. Posterior communicating artery aneurysm et al. Interpeduncular high-density spot in severe shearing presenting with haemorrhage into an arachnoid cyst. injury. Surg Neurol 1990; 34(1):30–8. J Neurol Neurosurg Psychiatry 1998; 64:558–60. 287 Dowling G, Curry B. Traumatic basal subarachnoid 271 Alliez B, Du Lac P, Trabulsi R. Anevrysme extra-cranien de hemorrhage. Report of six cases and review of the l’artère cerebelleuse postéro- inférieure. Une observation. literature. Am J Forensic Med Pathol 1988; 9(1):23–31. Neurochirurgie 1990; 36(2):137–40. 288 Jenkins A, Hadley DM, Teasdale GM, Condon B, 272 Hijdra A, van Gijn J. Early death from rupture of an Macpherson P, Patterson J. Magnetic resonance imaging intracranial aneurysm. J Neurosurg 1982; 57(6):765–8. of acute subarachnoid hemorrhage. J Neurosurg 1988; 273 Weir B, Myles T, Kahn M, Maroun F, Malloy D, Benoit B 68(5):731–6. et al. Management of acute subdural hematomas 289 Patel MR, Edelman RR, Warach S. Detection of hyperacute from aneurysmal rupture. Can J Neurol Sci 1984; primary intraparenchymal hemorrhage by magnetic 11(3):371–6. resonance imaging. Stroke 1996; 27:2321–4. 274 Kamiya K, Inagawa T, Yamamoto M, Monden S. Subdural 290 Mitchell P, Wilkinson ID, Hoggard N, Paley MNJ, Jellinek hematoma due to ruptured intracranial aneurysm. Neurol DA, Powell T et al. Detection of subarachnoid haemorrhage Med Chir Tokyo 1991; 31(2):82–6. with magnetic resonance imaging. J Neurol Neurosurg 275 Ohkuma H, Shimamura N, Fujita S, Suzuki S. Acute Psychiatry 2001; 70:205–11. subdural hematoma caused by aneurysmal rupture: 291 Noguchi K, Ogawa T, Inugami A, Toyoshima H, incidence and clinical features. Cerebrovasc Dis 2003; Okudera T, Uemura K. MR of acute subarachnoid 16:171–3. hemorrhage: a preliminary report of ﬂuid-attenuated 276 Inamasu J, Saito R, Nakamura Y, Ichikizaki K, Suga S, inversion-recovery pulse sequences. Am J Neuroradiol Kawase T et al. Acute subdural hematoma caused by 1994; 15(10):1940–3. ruptured cerebral aneurysms: diagnostic and therapeutic 292 Chrysikopoulos H, Papanikolaou N, Pappas J, Papandreou pitfalls. Resuscitation 2002; 52:71–6. A, Roussakis A, Vassilouthis J et al. Acute subarachnoid 277 Koerbel A, Ernemann U, Freudenstein D. Acute subdural haemorrhage: detection with magnetic resonance haematoma without subarachnoid haemorrhage caused by imaging. Br J Radiol 1996; 69:601–9. rupture of an internal carotid artery bifurcation aneurysm: 293 Wiesmann M, Mayer TE, Yousry I, Medele R, Hamann GF, case report and review of literature. Br J Radiol 2005; Brückmann HM. Detection of hyperacute subarachnoid 78:646–50. hemorrhage of the brain by using magnetic resonance 278 Pritz MB, Kaufman JK. Ruptured middle cerebral artery imaging. J Neurosurg 2002; 96:684–9. aneurysm and bilateral chronic subdural hematomas. 294 Fiebach JB, Schellinger PD, Geletneky K, Wilde P, Meyer M, Surg Neurol 2001; 55:123–5. Hacke W et al. MRI in acute subarachnoid haemorrhage; 279 Clarke E, Walton JN. Subdural haematoma complicating ﬁndings with a standardised stroke protocol. intracranial aneurysm and angioma. Brain 1953; Neuroradiology 2004; 46:44–8. 76:378–404. 295 Stone JL, Crowell RM, Gandhi YN, Jafar JJ. Multiple 280 Schievink WI, Wijdicks EFM, Piepgras DG, Nichols DA, intracranial aneurysms: magnetic resonance imaging for Ebersold MJ. Perimesencephalic subarachnoid determination of the site of rupture. Report of a case. hemorrhage. Additional perspectives from four cases. Neurosurgery 1988; 23(1):97–100. Stroke 1994; 25(7):1507–11. 296 Atlas SW. MR imaging is highly sensitive for acute 281 Huttner HB, Hartmann M, Kohrmann M, Neher M, subarachnoid hemorrhage ...not! Radiology 1993; Stippich C, Hahnel S et al. Repeated digital substraction 186(2):319–22. angiography after perimesencephalic subarachnoid 297 Matsumura K, Matsuda M, Handa J, Todo G. Magnetic hemorrhage? J Neuroradiol 2006; 33:87–9. resonance imaging with aneurysmal subarachnoid .. ..

9781405127660_4_009.qxd 10/13/07 11:03 AM Page 499 References 499 hemorrhage: comparison with computed tomography 312 Hirai T, Korogi Y, Ono K, Murata Y, Suginohara K, Omori T scan. Surg Neurol 1990; 34(2):71–8. et al. Preoperative evaluation of intracranial aneurysms: 298 Noguchi K, Ogawa T, Seto H, Inugami A, Hadeishi H, usefulness of intraarterial 3D CT angiography and Fujita H et al. Subacute and chronic subarachnoid conventional angiography with a combined unit: initial hemorrhage: diagnosis with ﬂuid-attenuated inversion- experience. Radiology 2001; 220:499–505. recovery MR imaging. Radiology 1997; 203(1):257–62. 313 Carvi y Nievas M, Haas E, Höllerhage HG, Drathen C. 299 Dreier JP, Sakowitz OW, Harder A, Zimmer C, Dirnagl U, Complementary use of computed tomographic Valdueza JM et al. Focal laminar cortical MR signal angiography in treatment planning for posterior fossa abnormalities after subarachnoid hemorrhage. subarachnoid hemorrhage. Neurosurgery 2002; 50:1283–8. Ann Neurol 2002; 52:825–9. 314 Velthuis BK, Van Leeuwen MS, Witkamp TD, Ramos LMP, 300 Mohamed M, Heasly DC, Yagmurlu B, Yousem DM. Berkelbach van der Sprenkel JW, Rinkel GJE. Computerized Fluid-attenuated inversion recovery MR imaging and tomography angiography in patients with subarachnoid subarachnoid hemorrhage: not a panacea. Am J Neuroradiol hemorrhage: from aneurysm detection to treatment 2004; 25:545–50. without conventional angiography. J Neurosurg 1999; 301 Imaizumi T, Chiba M, Honma T, Niwa J. Detection of 91:761–7. hemosiderin deposition by T2*-weighted MRI after 315 Matsumoto M, Sato M, Nakano M, Endo Y, Watanabe Y, subarachnoid hemorrhage. Stroke 2003; 34:1693–8. Sasaki T et al. Three-dimensional computerized 302 Quint DJ, Spickler EM. Magnetic resonance demonstration tomography angiography-guided surgery of acutely of vertebral artery dissection. Report of two cases. ruptured cerebral aneurysms. J Neurosurg 2001; 94:718–27. J Neurosurg 1990; 72(6):964–7. 316 Ross JS, Masaryk TJ, Modic MT, Ruggieri PM, Haacke EM, 303 Schwaighofer BW, Klein MV, Lyden PD, Hesselink JR. MR Selman WR. Intracranial aneurysms: evaluation by MR imaging of vertebrobasilar vascular disease. J Comput Assist angiography. Am J Roentgenol 1990; 155(1):159–65. Tomogr 1990; 14(6):895–904. 317 Okahara M, Kiyosue H, Yamashita M, Nagatomi H, Hata H, 304 Woimant F, Spelle L. Spontaneous basilar artery dissection: Saginoya T et al. Diagnostic accuracy of magnetic contribution of magnetic resonance imaging to diagnosis. resonance angiography for cerebral aneurysms in J Neurol Neurosurg Psychiatry 1995; 58:540. correlation with 3D-digital subtraction angiographic 305 White PM, Wardlaw JM, Easton V. Can noninvasive images – A study of 133 aneurysms. Stroke 2002; imaging accurately depict intracranial aneurysms? A 33:1803–8. systematic review. Radiology 2000; 217:361–70. 318 Mallouhi A, Felber S, Chemelli A, Dessl A, Auer A, Schocke 306 Chappell ET, Moure FC, Good MC. Comparison of M et al. Detection and characterization of intracranial computed tomographic angiography with digital aneurysms with MR angiography: comparison of volume- subtraction angiography in the diagnosis of cerebral rendering and maximum-intensity-projection algorithms. aneurysms: a meta-analysis. Neurosurgery 2003; 52:624–30. Am J Roentgenol 2003; 180:55–64. 307 Wintermark M, Uske A, Chalaron M, Regli L, Maeder P, 319 Ronkainen A, Puranen MI, Hernesniemi JA, Vanninen RL, Meuli R et al. Multislice computerized tomography Partanen PL, Saari JT et al. Intracranial aneurysms: MR angiography in the evaluation of intracranial aneurysms: angiographic screening in 400 asymptomatic individuals a comparison with intraarterial digital subtraction with increased familial risk. Radiology 1995; 195(1):35–40. angiography. J Neurosurg 2003; 98:828–36. 320 Raaymakers TWM, MARS Study Group. Aneurysms in 308 Villablanca JP, Hooshi P, Martin N, Jahan R, Duckwiler G, relatives of patients with subarachnoid hemorrhage: Lim S et al. Three-dimensional helical computerized frequency and risk factors. Neurology 1999; 53:982–8. tomography angiography in the diagnosis, 321 Cloft HJ, Joseph GJ, Dion JE. Risk of cerebral angiography characterization, and management of middle cerebral in patients with subarachnoid hemorrhage, cerebral artery aneurysms: comparison with conventional aneurysm, and arteriovenous malformation: a meta- angiography and intraoperative ﬁndings. J Neurosurg analysis. Stroke 1999; 30:317–20. 2002; 97:1322–32. 322 Hayakawa I, Watanabe T, Tsuchida T, Sasaki A. 309 White PM, Teasdale E, Wardlaw JM, Easton V. What is the Perangiographic rupture of intracranial aneurysms. most sensitive non-invasive imaging strategy for the Neuroradiology 1978; 16:293–5. diagnosis of intracranial aneurysms? J Neurol Neurosurg 323 Koenig GH, Marshall WH, Jr., Poole GJ, Kramer RA. Psychiatry 2001; 71:322–8. Rupture of intracranial aneurysms during cerebral 310 Van Gelder JM. Computed tomographic angiography for angiography: report of ten cases and review of the detecting cerebral aneurysms: implications of aneurysm literature. Neurosurgery 1979; 5(3):314–24. size distribution for the sensitivity, speciﬁcity, and 324 Saitoh H, Hayakawa K, Nishimura K, Okuno Y, Teraura T, likelihood ratios. Neurosurgery 2003; 53:597–605. Yumitori K et al. Rerupture of cerebral aneurysms during 311 Hashimoto H, Iida J, Hironaka Y, Okada M, Sakaki T. Use of angiography. Am J Neuroradiol 1995; 16(3):539–42. spiral computerized tomography angiography in patients 325 Velthuis BK, Van Leeuwen MS, Witkamp TD, Ramos LMP, with subarachnoid hemorrhage in whom subtraction Van der Sprenkel JWB, Rinkel GJE. Surgical anatomy angiography did not reveal cerebral aneurysms. of the cerebral arteries in patients with subarachnoid J Neurosurg 2000; 92:278–83. hemorrhage: comparison of computerized tomography .. ..

9781405127660_4_009.qxd 10/13/07 11:03 AM Page 500 500 Chapter 9 What caused this subarachnoid haemorrhage? angiography and digital subtraction angiography. 343 Logue V. Angiomas of the spinal cord: review of the J Neurosurg 2001; 95:206–12. pathogenesis, clinical features, and results of surgery. 326 Friedman AH, Drake CG. Subarachnoid hemorrhage J Neurol Neurosurg Psychiatry 1979; 42(1):1–11. from intracranial dissecting aneurysm. J Neurosurg 1984; 344 Savader SJ, Williams GM, Trerotola SO, Perler BA, Wang 60(2):325–34. MC, Venbrux AC et al. Preoperative spinal artery 327 Van der Meulen JHP, Weststrate W, van Gijn J, Habbema localization and its relationship to postoperative JD. Is cerebral angiography indicated in infective neurologic complications. Radiology 1993; 189:27–8. endocarditis? Stroke 1992; 23(11):1662–7. 345 D’Angelo V, Bizzozero L, Talamonti G, Ferrara M, Colombo 328 Ruelle A, Lasio G, Boccardo M, Gottlieb A, Severi P. Long- N. Value of magnetic resonance imaging in spontaneous term prognosis of subarachnoid hemorrhages of unknown extradural spinal hematoma due to vascular malformation: etiology. J Neurol 1985; 232(5):277–9. case report. Surg Neurol 1990; 34(5):343–4. 329 Spallone A, Ferrante L, Palatinsky E, Santoro A, Acqui M. 346 Mohsenipour I, Ortler M, Twerdy K, Schmutzhard E, Subarachnoid haemorrhage of unknown origin. Acta Attlmayr G, Aichner F. Isolated aneurysm of a spinal Neurochir (Wien) 1986; 80(1–2):12–17. radicular artery presenting as spinal subarachnoid 330 Juul R, Fredriksen TA, Ringkjob R. Prognosis in haemorrhage [letter]. J Neurol Neurosurg Psychiatry 1994; subarachnoid hemorrhage of unknown etiology. 57(6):767–8. J Neurosurg 1986; 64(3):359–62. 347 Schievink WI, Limburg M, Oorthuys JW, Fleury P, Pope 331 Suzuki S, Kayama T, Sakurai Y, Ogawa A, Suzuki J. FM. Cerebrovascular disease in Ehlers-Danlos syndrome Subarachnoid hemorrhage of unknown cause. type IV. Stroke 1990; 21(4):626–32. Neurosurgery 1987; 21(3):310–13. 348 Lee EK, Hecht ST, Lie JT. Multiple intracranial and systemic 332 Giombini S, Bruzzone MG, Pluchino F. Subarachnoid aneurysms associated with infantile-onset arterial hemorrhage of unexplained cause. Neurosurgery 1988; ﬁbromuscular dysplasia. Neurology 1998; 50:828–9. 22(2):313–16. 349 Cloft HJ, Kallmes DF, Kallmes MH, Goldstein JH, Jensen 333 Ciofﬁ F, Pasqualin A, Cavazzani P, Da Pian R. ME, Dion JE. Prevalence of cerebral aneurysms in patients Subarachnoid haemorrhage of unknown origin: clinical with ﬁbromuscular dysplasia: a reassessment. J Neurosurg and tomographical aspects. Acta Neurochir (Wien) 1989; 1998; 88:436–40. 97(1–2):31–9. 350 Román G, Fisher M, Perl DP, Poser CM. Neurological 334 du Mesnil de Rochemont R., Heindel W, Wesselmann C, manifestations of hereditary hemorrhagic teleangiectasis Kruger K, Lanfermann H, Ernestus RI et al. Nontraumatic (Rendu-Osler-Weber disease): report of two cases and subarachnoid hemorrhage: value of repeat angiography. review of the literature. Ann Neurol 1978; 4:130–44. Radiology 1997; 202:798–800. 351 Schievink WI, Mellinger JF, Atkinson JLD. Progressive 335 Urbach H, Zentner J, Solymosi L. The need for repeat intracranial aneurysmal disease in a child with progressive angiography in subarachnoid haemorrhage. Neuroradiology hemifacial atrophy (Parry-Romberg disease): case report. 1998; 40:6–10. Neurosurgery 1996; 38:1237–41. 336 Houben MP, Van Rooij WJ, Sluzewski M, Tijssen CC. 352 Schievink WI, Mokri B, Piepgras DG, Gittenberger-de [Subarachnoid hemorrhage without aneurysm on the Groot AC. Intracranial aneurysms and cervicocephalic angiogram: the value of repeat angiography]. Ned Tijdschr arterial dissections associated with congenital heart Geneeskd 2002; 146:804–8. disease. Neurosurgery 1996; 39:685–9. 337 Khan N, Schuknecht B, Yonekawa Y. Presentation and 353 Mercado R, López S, Cantú C, Sanchez A, Revuelta R, management of patients with initial negative 4-vessel Gómez-Llata S et al. Intracranial aneurysms associated with cerebral angiography in subarachnoid hemorrhage. unsuspected aortic coarctation – Report of three cases and Acta Neurochir Suppl 2002; 82:71–81. review of the literature. J Neurosurg 2002; 97:1221–5. 338 Inamasu J, Nakamura Y, Saito R, Horiguchi T, Kuroshima 354 Asaoka K, Houkin K, Fujimoto S, Ishikawa T, Abe H. Y, Mayanagi K et al. ‘Occult’ ruptured cerebral aneurysms Intracranial aneurysms associated with aortitis syndrome: revealed by repeat angiography: result from a large case report and review of the literature. Neurosurgery 1998; retrospective study. Clin Neurol Neurosurg 2003; 106:33–7. 42:157–60. 339 Di Lorenzo N, Guidetti G. Anterior communicating 355 Kayembe KN, Sasahara M, Hazama F. Cerebral aneurysms aneurysm missed at angiography: report of two cases and variations in the circle of Willis. Stroke 1984; treated surgically. Neurosurgery 1988; 23(4):494–9. 15(5):846–50. 340 Acciarri N, Padovani R, Pozzati E, Gaist G, Manetto V. 356 Horikoshi T, Akiyama I, Yamagata Z, Sugita M, Nukui H. Spinal cavernous angioma: a rare cause of subarachnoid Magnetic resonance angiographic evidence of sex-linked hemorrhage. Surg Neurol 1992; 37(6):453–6. variations in the circle of Willis and the occurrence of 341 Kandel EI. Complete excision of arteriovenous cerebral aneurysms. J Neurosurg 2002; 96:697–703. malformations of the cervical cord. Surg Neurol 1980; 357 Nagamine Y, Takahashi S, Sonobe M. Multiple intracranial 13(2):135–9. aneurysms associated with moyamoya disease: case report. 342 Swann KW, Ropper AH, New PF, Poletti CE. Spontaneous J Neurosurg 1981; 54(5):673–6. spinal subarachnoid hemorrhage and subdural hematoma. 358 Conway JE, Hutchins GM, Tamargo RJ. Lack of evidence Report of two cases. J Neurosurg 1984; 61(5):975–80. for an association between neuroﬁbromatosis type I and .. ..

9781405127660_4_009.qxd 10/13/07 11:03 AM Page 501 References 501 intracranial aneurysms: autopsy study and review of the 375 Venning MC, Burn DJ, Bashir SH, Deopujari CE, literature. Stroke 2001; 32:2481–5. Mendelow AD. Subarachnoid haemorrhage in Wegener’s 359 Van den Berg JSP, Limburg M, Hennekam RCM. Is Marfan granulomatosis, with negative four vessel angiography. syndrome associated with symptomatic intracranial Br J Neurosurg 1991; 5:195–8. aneurysms. Stroke 1996; 27:10–12. 376 Hentschel S, Toyota B. Intracranial malignant glioma 360 Conway JE, Hutchins GM, Tamargo RJ. Marfan syndrome presenting as subarachnoid hemorrhage. Can J Neurol Sci is not associated with intracranial aneurysms. Stroke 1999; 2003; 30:63–6. 30:1632–6. 377 Yonemitsu T, Niizuma H, Kodama N, Fujiwara S, Suzuki J. 361 Van den Berg JSP, Hennekam RCM, Cruysberg JRM, Acoustic neurinoma presenting as subarachnoid Steijlen PM, Swart J, Tijmes N et al. Prevalence of hemorrhage. Surg Neurol 1983; 20(2):125–30. symptomatic intracranial aneurysm and ischaemic stroke 378 Scotti G, Filizzolo F, Scialfa G, Tampieri D, Versari P. in pseudoxanthoma elasticum. Cerebrovasc Dis 2000; Repeated subarachnoid hemorrhages from a cervical 10:315–19. meningioma: case report. J Neurosurg 1987; 362 Escott EJ, Rubinstein D, Cajade-Law AG, Sze CI. Suprasellar 66(5):779–81. cavernous malformation presenting with extensive 379 Vilela P, Saraiva P, Goulao A. Intracranial angiolipoma as subarachnoid hemorrhage. Neuroradiology 2001; cause of subarachnoid haemorrhage: case report and 43:313–16. review of the literature. Neuroradiology 2005; 47:91–6. 363 Ohshima T, Endo T, Nukui H, Ikeda S, Allsop D, Onaya T. 380 Corriero G, Iacopino DG, Valentini S, Lanza PL. Cervical Cerebral amyloid angiopathy as a cause of subarachnoid neuroma presenting as a subarachnoid hemorrhage: case hemorrhage. Stroke 1990; 21:480–3. report. Neurosurgery 1996; 39:1046–9. 364 Takeda S, Yamazaki K, Miyakawa T, Onda K, Hinokuma K, 381 Caputi F, De Sanctis S, Gazzeri G, Gazzeri R. Neuroma of Ikuta F et al. Subcortical hematoma caused by cerebral the spinal accessory nerve disclosed by a subarachnoid amyloid angiopathy: does the ﬁrst evidence of hemorrhage hemorrhage: case report. Neurosurgery 1997; 41:946–50. occur in the subarachnoid space? Neuropathology 2003; 382 Collignon FP, Friedman JA, Atkinson JLD. Recurrent 23:254–61. intraventricular and subarachnoid hemorrhage due to an 365 Ishikawa T, Nakamura N, Houkin K, Nomura M. intracranial schwannoma: case illustration. J Neurosurg Pathological consideration of a ‘blister-like’ aneurysm at 2002; 96:377. the superior wall of the internal carotid artery: case report. 383 Parmar H, Pang BC, Lim CC, Chng SM, Tan KK. Spinal Neurosurgery 1997; 40:403–5. schwannoma with acute subarachnoid hemorrhage: a 366 Abe M, Tabuchi K, Yokoyama H, Uchino A. Blood diagnostic challenge. Am J Neuroradiol 2004; 25:846–50. blisterlike aneurysms of the internal carotid artery. 384 Inci S, Bozkurt G, Gulsen S, Firat P, Ozgen T. Rare J Neurosurg 1998; 89:419–24. cause of subarachnoid hemorrhage: spinal meningeal 367 Hochberg FH, Fisher CM, Roberson GH. Subarachnoid carcinomatosis: case report. J Neurosurg Spine 2005; hemorrhage caused by rupture of a small superﬁcial artery. 2:79–82. Neurology 1974; 24(4):319–21. 385 Montinaro A, Cantisani P, Punzi F, D’Agostino A. Cauda 368 Marushima A, Yanaka K, Matsuki T, Kojima H, Nose T. equina melanoma presenting with subarachnoid Subarachnoid hemorrhage not due to ruptured aneurysm hemorrhage: a case report. J Neurosurg Sci 2004; 48:139–41. in moyamoya disease. J Clin Neurosci 2006; 13:146–9. 386 Berlis A, Schumacher M, Spreer J, Neumann HP, van V, V. 369 Kim CH, Kim HJ. Cervical subarachnoid ﬂoating cavernous Subarachnoid haemorrhage due to cervical spinal cord malformation presenting with recurrent subarachnoid haemangioblastomas in a patient with von Hippel-Lindau haemorrhage. J Neurol Neurosurg Psychiatry 2002; 72:668. disease. Acta Neurochir (Wien) 2003; 145:1009–13. 370 Chehrenama M, Zagardo MT, Koski CL. Subarachnoid 387 Kallmes DF, Lanzino G, Dix JE, Dion JE, Do H, Woodcock hemorrhage in a patient with Lyme disease. Neurology RJ et al. Patterns of hemorrhage with ruptured posterior 1997; 48(2):520–3. inferior cerebellar artery aneurysms: CT ﬁndings in 44 371 Nakasu S, Kaneko M, Matsuda M. Cerebral aneurysms cases. Am J Roentgenol 1997; 169:1169–71. associated with Behcet’s disease: a case report. J Neurol 388 Lewis SB, Chang DWJ, Peace DA, LaFrentz PJ, Day AL. Neurosurg Psychiatry 2001; 70:682–4. Distal posterior inferior cerebellar artery aneurysms: 372 Munn EJ, Alloway JA, Difﬁn DC, Arroyo RA. Polyarteritis clinical features and management. J Neurosurg 2002; with symptomatic intracerebral aneurysms at initial 97:756–66. presentation. J Rheumatol 1998; 25:2022–5. 389 Horie N, Takahashi N, Furuichi S, Mori K, Onizuka M, 373 Takahashi JC, Sakai N, Iihara K, Sakai H, Higashi T, Kogure Tsutsumi K et al. Ruptured aneurysm at the choroidal S et al. Subarachnoid hemorrhage from a ruptured anterior branch of the posterior inferior cerebellar artery: a case cerebral artery aneurysm caused by polyarteritis nodosa: report and review of the literature. Surg Neurol 2003; case report. J Neurosurg 2002; 96:132–4. 60:540–4. 374 Calvo-Romero JM, del Carmen Bonilla-Gracia M, 390 Oana K, Murakami T, Beppu T, Yamaura A, Kanaya H. Bureo-Dacal P. Churg-Strauss syndrome presenting as Aneurysm of the distal anterior inferior cerebellar artery spontaneous subarachnoid haemorrhage. Clin Rheumatol unrelated to the cerebellopontine angle: case report. 2002; 21:261–3. Neurosurgery 1991; 28(6):899–903. .. ..

9781405127660_4_009.qxd 10/13/07 11:03 AM Page 502 502 Chapter 9 What caused this subarachnoid haemorrhage? 391 van Rybroek JJ, Moore SA. Sudden death from choroid associated with intraventricular hemorrhage and plexus vascular malformation hemorrhage: case report tuberculous peritonitis. Jpn J Med 1991; 30(3):233–7. and review of the literature. Clin Neuropathol 1990; 400 Pallud J, Nataf F, Roujeau T, Roux FX. Intraventricular 9(1):39–45. haemorrhage from a renal cell carcinoma pituitary 392 Kataoka K, Taneda M. Angiographic disappearance of metastasis. Acta Neurochir (Wien) 2005; 147:1003–4. multiple dural arteriovenous malformations: case report. 401 Poon TP, Solis OG. Sudden death due to massive J Neurosurg 1984; 60(6):1275–8. intraventricular hemorrhage into an unsuspected 393 Irie F, Fujimoto S, Uda K, Toyoda K, Hagiwara N, Inoue T ependymoma. Surg Neurol 1985; 24(1):63–6. et al. Primary intraventricular hemorrhage from dural 402 Bosnjak R, Derham C, Popovic M, Ravnik J. Spontaneous arteriovenous ﬁstula. J Neurol Sci 2003; 215:115–18. intracranial meningioma bleeding: clinicopathological 394 Jabbour R, Taher A, Shamseddine A, Atweh SF. Moyamoya features and outcome. J Neurosurg 2005; 103:473–84. syndrome with intraventricular hemorrhage in an adult 403 Pascual J, Diez C, Carda JR, Vazquez Barquero A. with factor V Leiden mutation. Arch Neurol 2005; Intraventricular haemorrhage complicating a brain 62:1144–6. abscess. Postgrad Med J 1987; 63(743):785–7. 395 Masson C, Martin N, Masson M, Cambier J. Hémorragie 404 Wong CW, Ho YS. Intraventricular haemorrhage and intraventriculaire après endarteriectomie carotidienne. hydrocephalus caused by intraventricular parasitic Role des suppléances de type moyamoya. Rev Neurol (Paris) granuloma suggesting cerebral sparganosis. Acta Neurochir 1986; 142(8–9):716–19. (Wien) 1994; 129(3–4):205–8. 396 Hamada J, Hashimoto N, Tsukahara T. Moyamoya disease 405 Imanse J, Vanneste JAL. Intraventricular hemorrhage with repeated intraventricular hemorrhage due to following amphetamine abuse. Neurology 1990; aneurysm rupture: report of two cases. J Neurosurg 1994; 40(8):1318–19. 80(2):328–31. 406 Moriya F, Hashimoto Y. A case of fatal hemorrhage 397 Gates PC, Barnett HJM, Vinters HV, Simonsen RL, Siu K. in the cerebral ventricles following intravenous use of Primary intraventricular hemorrhage in adults. Stroke methamphetamine. Forensic Sci Int 2002; 129:104–9. 1986; 17(5):872–7. 407 Pﬁster HW, Von Rosen F. Severe intraventricular 398 Stieltjes N, Calvez T, Demiguel V, Torchet MF, Briquel ME, haemorrhage shown by computed tomography as an Fressinaud E et al. Intracranial haemorrhages in French unusual manifestation of Wernicke’s encephalopathy. haemophilia patients (1991–2001): clinical presentation, J Neurol Neurosurg Psychiatry 1995; 59:555–6. management and prognosis factors for death. Haemophilia 408 Van der Wee N, Rinkel GJE, Hasan D, van Gijn J. Detection 2005; 11:452–8. of subarachnoid haemorrhage on early CT scans: is lumbar 399 Tsubota A, Shishiba Y, Shimizu T, Ozawa Y, Sawano S, puncture still indicated after a negative scan. J Neurol Yamada S. Masked Cushing’s disease in an aged man Neurosurg Psychiatry 1995; 58:357–359. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 503 10 A practical approach to the management of stroke and transient ischaemic attack patients 10.1 Aims of treatment 503 10.2 What is this patient’s prognosis? 505 10.3 Delivering an integrated management plan 518 10.4 Some difficult ethical issues 531 This chapter introduces the general principles of manag- A revision of this original classiﬁcation, the WHO ing patients with stroke. Because treatment is aimed at International Classiﬁcation of Functioning, Disabilities improving the patient’s outcome, the chapter includes a and Health substituted some new terms (Table 10.1) for section on the early prognosis of stroke and the factors old ones to emphasize positive aspects (i.e. activity, not that may help predict the progress of individual patients disability; participation, not handicap) and to highlight over the ﬁrst year. It also introduces a model for treat- the important ‘contextual’ factors – e.g. personal experi- ing patients that avoids the pitfalls of the traditional ences, physical and social environment – that inﬂuence approach, which splits treatment artiﬁcially into acute the impact of disease, at each level, on the individual. 2,3 care, rehabilitation and continuing care. However, we ﬁnd the original version easier to relate to our patients’ condition and therefore will refer to that version in the following sections. Although not included in the WHO classiﬁcations, quality of life is obviously an important aspect of a pa- 10.1 Aims of treatment tient’s outcome. However, there is no generally accepted deﬁnition of quality of life, and it is therefore not surpris- ing that it is difﬁcult to measure (section 17.12.5). The aims of treatment can be summarized as optimizing the patient’s chance of surviving and minimizing the The consequences of a stroke must be considered impact of the stroke and any recurrent vascular events at ﬁve levels: pathology, impairment, disability, on the patient and carers. In minimizing the impact handicap and quality of life. of the stroke, one has to think not just about the short-term effects of the stroke in causing the patient’s The most obvious effects of a stroke are physical, but neurological impairments, but also about its effect on in many situations these may not be as important as the patient’s function (i.e. disability) and role in society the cognitive, psychological, social and even ﬁnancial (i.e. handicap). Therefore, it is useful to consider the consequences. Thus, treatment that aims to minimize consequences of a stroke in terms of the original World the impact of a stroke on patients and their carers must Health Organization (WHO) International Classiﬁcation be directed at all of these various problems. of Impairments, Disabilities and Handicaps (ICIDH). 1 10.1.1 Aspects of treatment Each patient has a unique blend of pathologies, impair- Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, ments, disabilities and handicaps. Therefore it follows M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published that treatment must be preceded by a comprehensive 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0. assessment and then tailored to that individual patient. 503 ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 504 504 Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients Table 10.1 Levels within the World Health Organization (WHO) International Classiﬁcation of Impairments, Disabilities and Handicaps (with some new terms introduced in the WHO Assessment International Classiﬁcation of Functioning, Disabilities and Goal setting Health). Monitoring & re-assessment Multidisciplinary Pathology: the underlying pathological substrate of the teamwork stroke, e.g. ischaemic stroke due to embolic occlusion of Problem Education solving a middle cerebral artery from thrombus in the left atrium, & training resulting from atrial ﬁbrillation due to ischaemic heart Physical disease. Speciﬁc medical and surgical treatments (e.g. therapy thrombolytic or neuroprotective drugs) are directed at this level of the disease process (Chapters 12–15). Impairment: any loss or abnormality of speciﬁc Fig. 10.1 The complex process called ‘rehabilitation’. psychological, physiological, or anatomical structure or function caused by the stroke (e.g. muscle weakness or Having referred the patient to one or more therapists, a spasticity, loss of sensation, aphasia). Physical therapies, physician then mistakenly believes that ‘rehabilitation’ such as physiotherapy or electromyographic biofeedback, has been organized. This is far too simplistic (Fig. 10.1). are directed at this level (section 11.21). Although there is no universally accepted deﬁnition of Disability (activity): any restriction or lack (resulting from rehabilitation, most people would view it as a ‘goal- an impairment) of ability to perform an activity in the manner or within the range considered normal for a orientated’ process aimed at minimizing the functional human being (e.g. inability to walk, wash, feed, etc.) due consequences of the stroke, minimizing the impact of to the stroke. Physical therapies are also used to try to the stroke on the lives of the patient and any carers, and reduce the disability related to impairments. maximizing their autonomy. If we include in our deﬁni- Handicap (participation): the disadvantage for a given tion all those components of care that have these aims, it individual, resulting from an impairment or disability, is apparent that rehabilitation must embrace most aspects that limits or prevents the fulﬁlment of a role (depending of care, ranging from the acute medical treatment through on age, sex, social and cultural factors) for that individual to making alterations to the patient’s home prior to hos- – e.g. inability to continue the same job. Although more pital discharge and providing support later on. Achiev- difﬁcult to deﬁne and measure than the other levels of ing the best possible outcome for the patient requires a disease, handicap is probably the level that best reﬂects broad approach rather than one that just focuses on the the patient’s and carer’s perspective. Many aspects of treatment will impact on handicap, but occupational primary lesion, or just on the resulting impairments. therapy and social work are those most obviously aimed at inﬂuencing this level. Rehabilitation is not synonymous with physical therapies such as physiotherapy or occupational therapy – it is a far more complex process including Traditionally, the discussion of the treatment of stroke is assessment, goal-setting, physical therapies, split into sections on: general treatment in the acute reassessment and teamwork. phase; acute medical and surgical treatments; secondary prevention; rehabilitation; and continuing care. How- When the problem is thought of in this way, it becomes ever, this structure does not reﬂect the need for an artiﬁcial – and perhaps even harmful – to separate stroke integrated approach to the management of the patient, management into acute care, secondary prevention, even though remnants of this structure remain in this rehabilitation and continuing care. All aspects are going book. For example, patients may develop acute problems on simultaneously. To compound the problem, these (e.g. pneumonia, pulmonary embolism or urinary tract separate components of care may even be provided by infection) or recurrent vascular events at any stage in different staff in different institutions, which leads to a their illness, quite often during what is commonly called breakdown in communication and lack of continuity of 4 rehabilitation. Conversely, certain aspects of rehabilita- care. Often, in this modular system of care, one encoun- tion, such as teamwork and early mobilization, are just as ters a patient who is ‘waiting for rehabilitation’, i.e. the important on the day of stroke onset as they are later on. patient is in the department that normally deals with The term ‘rehabilitation’ means different things to dif- acute stroke patients but there is no immediate place ferent people. Unfortunately, to many physicians who in the rehabilitation facility and so the patient is not are responsible for the care of stroke patients, the term is progressing. Conversely, one comes across patients in synonymous with physical therapy (e.g. physiotherapy, a ‘rehabilitation setting’ who have developed acute med- occupational therapy, and speech and language therapy). ical problems (e.g. epilepsy or chest pain) and who are .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 505 10.2 What is this patient’s prognosis? 505 Assessment of each other and cycling at different rates, with each Swallowing having some inﬂuence on the others. This model of management applies equally well to acute general care, rehabilitation and continuing care. Short-loop 10.1.3 A guide to the following sections on Intervention Identify problem management Restrict oral intake Give intravenous fluids We have tried to reﬂect our integrated approach in the structure of the following discussion on management, Risk of Unable to aspiration swallow safely which is divided into several sections: Long-loop What is this patient’s early prognosis? Section 10.2 deals with the prognosis of stroke over the ﬁrst year or so, Intervention with respect to survival and function in groups of Teach compensatory patients, and in individual patients. techniques Delivering an integrated management plan: Section 10.3 Fig. 10.2 ‘Short-loop’ and ‘long-loop’ problems. deals with general assessment of the patient, the role of the stroke team and its members, and problem- denied quick access to the necessary facilities or expertise orientated and goal-orientated care. to ensure optimum management of the problems. Some difﬁcult ethical dilemmas: In section 10.4, we discuss some of the ethical dilemmas that arise in treating We should abandon the arbitrary division of treatment stroke patients. into acute and rehabilitation phases and adopt an What are the patient’s problems? A problem-orientated integrated, problem- and goal-orientated approach. approach to management: Chapter 11 deals with the problems that may occur after stroke, their assessment, and interventions that may help to prevent or solve 10.1.2 An integrated, problem-orientated and these problems. goal-orientated approach Speciﬁc medical and surgical treatments in the acute phase: The patient’s general management – as distinct from the Chapters 12–14 deal with the pathophysiology of speciﬁc treatment of the stroke pathology (Chapters acute stroke and the drug and surgical treatments that 12–14) – is primarily aimed at anticipating and prevent- aim to reduce the severity of brain injury. ing potential problems, and solving existing ones that Preventing recurrent stroke and other serious vascular events: are identiﬁed at various stages of the illness. One can Chapter 15 deals with speciﬁc interventions to prevent think of management in terms of many interwoven intracerebral haemorrhage while Chapter 16 com- cycles or loops (Fig. 10.2). The assessment of a problem, pletes the description of the prognosis of stroke by or potential problem, includes not only detecting and focusing on the early and later risks of further stroke perhaps measuring it, but also considering its likely and other serious vascular events, before moving on to cause and prognosis. This assessment may often have to describe various strategies to reduce these risks. include the patient’s or carer’s expectations or wishes Organizing stroke services: Finally, Chapter 17 focuses (section 10.3.4). Furthermore, assessment is not just a on the organizational issues that are important when ‘once only’ activity, but one that should be repeated trying to deliver all these various aspects of treatment throughout the illness so that management is tailored to to large numbers of stroke patients as efﬁciently and the patient’s needs as they change and evolve. equitably as possible. For some problems, this cycle can be completed in a few minutes (e.g. an obstructed airway is a short-loop problem), while for others, the cycle might take weeks to complete (e.g. depression is a long-loop problem). Quite often a problem such as dysphagia will demand both an 10.2 What is this patient’s prognosis? immediate intervention – e.g. stopping oral intake and giving ﬂuids and perhaps nutrition by an alternative route – and longer-term interventions, such as providing 10.2.1 Introduction retraining to compensate for swallowing impairment (sections 11.17 and 11.19). Thus, in reality, the general It is useful to try to predict the outcome of individual management involves many such cycles layered on top patients, because this may enable one to: .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 506 506 Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients Table 10.2 Methodological features that are important in assessing a study of prognosis after stroke. 72 Were the patients identiﬁed at an early and uniform point in the example, patients with a good recovery may be more course of their disease, and were diagnostic criteria, disease mobile or at work and therefore more difﬁcult to follow up, severity, comorbidity and demographic details for inclusion whilst patients may not be followed up because they have clearly speciﬁed? died. Therefore, the effect of incomplete follow up on If they were, then this is called an ‘inception cohort’. When prognosis is difﬁcult to predict. applying data from a study of prognosis, it is important to Were objective outcome criteria developed and used, and were the consider whether the patients studied were similar to one’s criteria reproducible and accurate? own patients. To make sense of prognostic data, it is important to know Was the referral pattern described? what the authors meant by terms such as ‘recurrent stroke’ Did the study avoid the following: or ‘independent’, so that one can apply the data to one’s – ‘referral ﬁlter bias’, which occurs when an inception cohort own patients. It is also important that the criteria were is assembled on selected cases that are not representative of applied consistently. all cases occurring in the population. This is a particular Was outcome assessment blind? problem with specialist centres, which attract unusual cases In other words, were diagnostic suspicion bias and (centripetal bias) or admit or track interesting cases expectation bias avoided in the assessment of patient (popularity bias). outcomes? If the observer has a preconceived view that a – ‘diagnostic access bias’, which occurs if cases are deﬁned by particular baseline factor is likely to be related to a particular technology (e.g. intracerebral haemorrhage by CT outcome, knowledge of the presence or absence of that scanning), but the patient’s access to the technology is factor at the time of follow-up may bias that observer. inﬂuenced by factors such as their wealth, which may affect Was adjustment for extraneous prognostic factors carried out? their outcome. Where authors relate certain baseline factors to the likelihood Was consent bias avoided? of speciﬁc outcomes, it is important that they should allow If inclusion in a cohort requires the patient to give explicit for other baseline factors. The most common example of consent then those from whom consent could not be this is age, which partly explains the observed relationships obtained will be excluded. This may include those who between other factors, e.g. atrial ﬁbrillation and early refuse consent or more often those in whom there were death. Before applying predictive equations to one’s own inadequate research resources to request consent. Patients patients, it is important for the equation to have been from whom consent could not be obtained may differ tested on an independent test cohort other than the one systematically from those included and thus their exclusion from which it was developed. may introduce bias in the assessment of prognosis. 73 Was the study prospective or retrospective? Was complete follow-up achieved? In general, prospective studies provide more reliable data Were all patients who were entered into the study accounted than retrospective ones, because cases and events during for, and was their clinical status known at the ﬁnal follow- the follow-up can be deﬁned using strict criteria, complete up? Patients who are lost to follow-up may be data are more likely to be available, and these studies are systematically different from those who are not. For less prone to bias. CT, computerized tomography. • have more informed discussions with the patient wasteful to use resources on patients who will make and/or their carers; a good recovery without any intervention at all. Of • set more appropriate short-term and long-term goals course, by adopting this approach, one must avoid (section 10.3.3); self-fulﬁlling prophecies – i.e. if one withdraws treat- • weigh the potential risks and beneﬁts of treatment ment from a patient, he or she may do badly because of options (e.g. one might reserve a particularly hazardous the lack of input. but nevertheless effective treatment for patients in Before considering how to predict an individual’s whom the prognosis is poor); prognosis, we shall describe the prognosis of the ‘aver- • plan treatment and make early decisions about later age’ patient, i.e. the outcome of an unselected cohort of discharge and long-term placement to optimize the stroke patients. Here, the prognosis with respect to sur- efﬁciency of the service; vival and overall functional outcome is described, since • make rationing decisions where resources are limited. this is relevant to all aspects of treatment. The prognosis Thus, if a particular patient is very unlikely to make a for particular individual impairments, disabilities and good recovery, one could divert resources from that handicaps is dealt with in the appropriate sections of patient to another with a better prognosis who may Chapter 11, while that relating to the risk of late death, gain more from the interventions available. It is also recurrent stroke and other vascular events is dealt with .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 507 10.2 What is this patient’s prognosis? 507 in Chapter 16. The prognosis of subarachnoid hae- be improving over time in some populations, although morrhage is described in detail in Chapter 14. it is unclear whether this is due to improved health of the population, changing severity of stroke, improved treatment, or methodological factors such as improved 10.2.2 Collecting reliable information about 6–8 detection of milder strokes (section 18.2). prognosis If information about the prognosis of stroke is to be 10.2.3 Prognosis for death useful, it must have been collected using sound methods that minimize bias and maximize precision, accuracy The risks of dying within the ﬁrst 7 days or 30 days after and generalizability (Table 10.2). a ﬁrst-ever-in-a-lifetime stroke are about 10% and 20%, respectively (Table 10.3). The risk of dying in the years after a stroke remains higher than for stroke-free indi- Prognosis or natural history? 9 viduals (Fig. 10.3). Patients with haemorrhagic stroke, It is important to distinguish between these two terms. either intracerebral or subarachnoid, have a much higher Natural history refers to the untreated course of an illness early risk of dying than those with ischaemic stroke from its onset, while prognosis refers to the probability although this might be partly explained by studies of a particular outcome occurring either in an individual missing small intracerebral haemorrhages due to delays or a group of patients over a deﬁned period of time after in brain imaging (Figs 5.1 and 10.4). Patients with major the disease is ﬁrst identiﬁed. The prognosis is likely to ischaemic strokes, i.e. total anterior circulation infarc- be inﬂuenced by any treatment given. Usually, but not tion (section 4.3.4), also have a very high early risk of always, the prognosis with treatment is better than the death (Table 10.3). natural history, but it may be worse. This section describes the prognosis of stroke. No data on the natural Causes of death history (strictly deﬁned) are available, because in most parts of the world, patients with stroke are usually given Knowing the causes of these early deaths is import- some treatment, and in those places where minimal ant if one is to prevent them (Fig. 10.5). In the ﬁrst or no treatment is given, no studies of natural history few days after stroke, most patients who die generally have been reported. Even admission to a hospital, even do so as a result of the direct effects of brain damage. 10 without any medical or physical therapy, is an inter- In brainstem strokes, the respiratory centre may be vention and could be regarded as ‘treatment’ which affected by the stroke itself, while in supratentorial may inﬂuence outcome. ischaemic or haemorrhagic stroke, dysfunction of the 100 Sources of prognostic data No published study of prognosis after stroke completely 80 Expected fulﬁls all the criteria summarized in Table 10.2. We have mainly used data from two studies performed in Oxford- 60 shire, UK, which used similar methods which at least Per cent surviving Average risk of death in partly met these criteria. The Oxfordshire Community 40 30 day survivors = 8.7%/year Stroke Project collected data during the 1980s and the 95% confidence Oxford Vascular Study (OXVASC) over the last few 20 interval years. 5,6 Other methodologically sound studies come to broadly similar conclusions, although to compare them 0 0 1 2 3 4 5 directly is difﬁcult because of their different methods, Years their varying styles of reporting, and because much of n = 675 467 399 274 182 92 the variation in prognosis can be accounted for by differ- Fig. 10.3 A Kaplan–Meier plot, showing the proportion of ences in case mix and by the play of chance as a result patients surviving at increasing intervals after a ﬁrst-ever-in-a- of relatively small sample sizes (section 17.12.6). Most lifetime stroke compared with the expected survival of people studies have included predominantly white patients of the same age and sex who have not had a stroke (data from managed in quite well-organized healthcare systems – so the Oxfordshire Community Stroke Project). The expected one must be careful in extrapolating the results to other survival was derived from all-cause mortality data for ethnic groups being cared for in different environments. Oxfordshire (1985) (reproduced with permission from There is some evidence that the prognosis of stroke may Dennis et al. 1993). 75 .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 508 508 Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients Table 10.3 Death after different (a) OCSP pathological types, and ischaemic Case fatality (%) stroke subtypes, of ﬁrst-ever-in-a-lifetime stroke. Data from (a) the Oxfordshire Type of stroke n 7 days 30 days 6 months 1 year Community Stroke Project (OCSP) performed in the 1980s and (b) OXVASC All strokes 675 12 19 27 31 performed in the same population since Subarachnoid haemorrhage 33 27 46 48 48 2000. Intracerebral haemorrhage 66 40 50 58 62 All ischaemic stroke 545 5 10 18 23 Total anterior circulation infarct 92 17 39 57 60 Partial anterior circulation infarct 186 2 4 11 16 Lacunar infarct 138 2 2 7 11 Posterior circulation infarct 129 5 7 14 19 (b) OXVASC Case fatality (%) Type of stroke n 7 days 30 days 6 months 1 year All strokes 345 11 15 22 24 Subarachnoid haemorrhage 22 41 45 45 45 Intracerebral haemorrhage 33 42 52 58 61 All ischaemic stroke 290 5 9 16 18 Total anterior circulation infarct 22 41 50 64 64 Partial anterior circulation infarct 122 1 4 16 19 Lacunar infarct 80 0 0 1 3 Posterior circulation infarct 66 8 15 18 20 100 Neurological Immobility Recurrence Ischaemic stroke Cardiac Non-vascular Uncertain 80 Subarachnoid 100 Per cent surviving 60 haemorrhage 80 40 60 haemorrhage 20 Intracerebral Per cent 40 20 0 0 12 3 4 5 0 Years <7 days 7–30 30 days– 6 months– >1 year days 6 months 1 year Fig. 10.4 A Kaplan–Meier plot, showing the proportion of patients surviving after a ﬁrst-ever-in-a-lifetime ischaemic Interval stroke (n = 545), intracerebral haemorrhage (n = 66) and Fig. 10.5 Histogram showing the proportion of patients dying subarachnoid haemorrhage (n = 33) (reproduced with from different causes at increasing intervals after a ﬁrst-ever-in- permission from Dennis et al. 1993). 75 a-lifetime stroke (from Bamford et al.; 10 Dennis et al. 1993). 75 brainstem results from displacement and herniation are probably due to coexisting cardiac pathology, or of oedematous supratentorial brain tissue (Figs 12.8 perhaps very rarely cardiac complications of the stroke and 12.9). Deaths occurring within 1–2 h of onset are (section 11.2.3). very unusual in ischaemic stroke, because it takes time Death within a few hours of stroke onset can occur for cerebral oedema to develop. Almost all such very with intracerebral or subarachnoid haemorrhage, or early deaths after stroke result from intracranial hae- 5 morrhage of some sort. The very few sudden deaths rarely with massive brainstem infarction. .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 509 10.2 What is this patient’s prognosis? 509 Having survived the ﬁrst few days, patients may then Dead Dependent Independent develop various potentially fatal complications of immo- bility, the most common being pneumonia (section 11.12) 100 and pulmonary embolism (section 11.13). In addition, 80 pressure ulcers (section 11.16), dehydration (section 11.18.1) with renal failure, and urinary tract infection 60 (section 11.12) may cause death where there is little if Per cent any basic care. Because some strokes occur in the context 40 of other serious conditions, e.g. myocardial infarction (section 7.10), cardiac failure (section 6.5.11) and cancer 20 (section 7.12), some early deaths can, at least in part, be 0 attributed to these underlying problems. Also, because Pre-stroke 1 month 6 months 1 year 2 years the risk of stroke recurrence is highest early after the ﬁrst Interval stroke – about 20% in the ﬁrst year (section 16.2) – some Fig. 10.6 Histogram showing the proportion of patients with patients will die from the direct or indirect effects of a different outcomes (i.e. dead, dependent (modiﬁed Rankin 3, 11 recurrent stroke . This is most likely in patients with 4 or 5) or independent (modiﬁed Rankin 0, 1 or 2) in activities aneurysmal subarachnoid haemorrhage (section 14.4.1), of daily living) at increasing intervals after their ﬁrst-ever-in-a- in whom the recurrence or rebleed rate is about 30% lifetime stroke (unpublished data from the Oxfordshire (without intervention), accounting for the majority of Community Stroke Project). deaths (50% of the total) in the ﬁrst 30 days. 12 caps will be discussed in the speciﬁc sections dealing 10.2.4 Prognosis for dependency in everyday with their treatment (Chapter 11), but it is relevant here activities to discuss the general pattern of recovery after stroke. Stroke often leaves surviving patients with neurological impairments that prevent them from performing every- 10.2.5 Patterns of recovery day activities and therefore dependent on others. Fig- ures 10.6 and 10.7 show the proportions of survivors Patients who survive an acute stroke almost always who are independent or dependent in everyday activities improve to a greater or lesser extent. Improvement is at various times after a ﬁrst-ever-in-a-lifetime stroke and reﬂected not just in a reduction in the neurological in strokes of different pathologies and subtypes. Other impairments but also in any resulting disability and studies have produced similar data. 13 It is likely that a handicap. The overall ‘pattern of recovery’ reﬂects sev- greater proportion of patients will become dependent eral processes superimposed upon each other. 14 In the after recurrent strokes. Details of the prognosis with ﬁrst few days after a stroke, ischaemic neurones that were respect to particular impairments, disabilities and handi- not irreversibly damaged during the primary event (i.e Dead Dependent Independent 100 Fig. 10.7 Histogram showing the proportion of patients with different outcomes (i.e. dead, dependent (modiﬁed 80 Rankin 3, 4 or 5) or independent (modiﬁed Rankin 0, 1 or 2) in activities of 60 daily living) 1 year after their ﬁrst-ever-in- Per cent a-lifetime ischaemic stroke (n = 545) and its clinical subtypes (TACI, total anterior 40 circulation infarction; PACI, partial anterior circulation infarction; LACI, 20 lacunar infarction; POCI, posterior circulation infarction), intracerebral 0 haemorrhage (ICH; n = 66) and All strokes ICH SAH All ischaemic TACI PACI LACI POCI subarachnoid haemorrhage (SAH; strokes n = 33) (from Bamford et al. 1990). 10 Type of stroke .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 510 510 Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients those in the ischaemic penumbra), may start to function been irreversibly damaged – might also explain some because of improved blood supply, reversal of metabolic of the later improvement (Fig. 10.8). However, much problems, or reduction of cerebral oedema (section of the later recovery with respect to disability and handi- 12.1.5). Resolution of diaschisis (section 5.6) is another cap is probably due to adaptive changes – i.e. patients explanation for early recovery, although this mech- learn techniques to compensate for their remaining anism has not been well established. Neuroplasticity – impairments, and their environment is altered to the process by which other intact areas of the brain maximize their autonomy. Although it is difﬁcult to can take over some of the functions of those that have predict an individual patient’s functional outcome, the (a) Fig. 10.8 Functional magnetic resonance imaging (fMRI) studies of dynamic changes in patterns of brain activation associated with hand movement accompanying recovery from an ischaemic stroke. The panels show the blood oxygenation level-dependent (bold) fMRI activations in the cerebral cortex and cerebellum accompanying ﬂexion– extension of the digits of one hand at (b) about 1.5 Hz. These activation volumes show local areas of increased blood ﬂow with task-speciﬁc neuronal activation. The right hemisphere is represented on the left side of the images. Activation of the supplementary motor cortex (SMA), primary sensorimotor cortex (MC) and cerebellum are easily identiﬁed. (Also reproduced in colour in the plate section.) (Images courtesy of H. Johannsen-Berg, S. Pendlebury and P.M. Matthews, Centre for Functional Magnetic Resonance Imaging (c) of the Brain, University of Oxford.) From 2 weeks after the infarct (a) to 5 weeks after the infarct (b), there was functional improvement in the paretic left hand associated with increased activation of right hemisphere MC and decreased activation in SMA and the cerebellum. In contrast, movements of the unaffected right hand do not show clearly signiﬁcant changes in the patterns of brain activation between the 2-week (d) scan (c) and 5-week scan (d). .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 511 10.2 What is this patient’s prognosis? 511 following general points can be made to patients and them. For example, it is often said that language func- their relatives: tion continues to improve for a very long time after a • the rate of recovery is usually fastest in the ﬁrst few stroke, while recovery in arm function does not. This dif- weeks after the initial stroke; ferent perception may simply be due to patients being • functional improvement may continue, albeit at a slower acutely aware of even small differences in ﬂuency, while rate, for many months and in some patients for 1–2 years; they may only report an improvement in arm function if • the speed and completeness of recovery varies from they can perform a new function with their hand. More patient to patient and is relatively unpredictable, at research into the patterns of recovery after stroke that least in the ﬁrst few days and weeks after stroke onset. takes these points into account would be helpful. It is unclear whether patterns of recovery differ among The pattern of recovery varies among patients and different pathological types of stroke (haemorrhagic in individuals, and rarely follows that implied by vs ischaemic) 19–21 or subtypes of ischaemic stroke (lacu- grouped data. Only repeated assessments in individual nar vs cortical). Clinical experience suggests that recov- patients can indicate their own pattern of recovery. ery patterns do vary. For instance, we have seen late and dramatic improvements after intracerebral haemor- These generalizations are supported by our own ex- rhage, but there are few reliable research data to conﬁrm perience and also by data from studies in which stroke such observations. Although individual patients may patients’ functional abilities have been repeatedly tested continue to improve for a year or even two, the mean or over a period of time. 15–17 Figure 10.9 shows grouped median measure of function in a cohort of stroke patients and individual patient data during recovery measured peaks at about 6 months and then begins to slowly with the Barthel index. The chart shows grouped data decline. 17 Any decline is presumably the result of a com- which supports the idea that the ‘pattern of recovery’ bination of subsequent ageing, recurrent strokes, progres- follows an almost exponential trajectory. The individual sion of comorbidity, and perhaps withdrawal of physical patients within the cohort frequently have different pat- therapy, other services and supports (section 11.6). terns of recovery. Also, the apparent plateau in recovery after a few months may simply reﬂect the fact that the The shape of so-called recovery curves may reﬂect the tools used to measure the function are often ‘ordinal’ properties of the instrument used to measure function rather than ‘interval’ scales (section 17.12.5), and also as much as the patient’s rate of improvement. that there is a marked ‘ceiling effect’ – i.e. the measure is not sensitive to improvements at the upper end of the 10.2.6 Is this the prognosis of your patients? range of performance. 18 Therefore, the apparent differ- ences in the patterns and duration of recovery for differ- The prognostic data presented in this section come mainly ent impairments and disabilities may to some extent from two cohorts studies in Oxfordshire, UK performed reﬂect the characteristics of the tools used to measure about 20 years apart. 5–6 These prospectively registered 120 Fig. 10.9 Patterns of recovery after stroke. The graph (open circles) shows the changes in the individual functional 100 independence measures (FIM) measured weekly in a group of patients with a partial anterior circulation stroke (PACS). 80 The closed circles indicate the death of the FIM score patients. The solid lines represent the median FIM with interquartile ranges. 60 Although the grouped data suggest an almost exponential recovery, this pattern 40 was rarely observed in individual patients. (Reproduced with permission from Langhorne P, Murray GD, Stott DJ, 20 McAlpine C, MacDonald J. Proﬁles of recovery after stroke: their value in monitoring 051015 20 25 recovery and predicting outcome. Chief Scientists Ofﬁce report. K/RED/4/C284.) 76 Week .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 512 512 Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients all patients from a well-deﬁned population who had a that are likely to lead to death within a few hours, ﬁrst-ever-in-a-lifetime stroke. After assessment by a study or who are already living in a nursing home, may not neurologist, as soon after the stroke onset as possible, be admitted at all and therefore be underrepresented patients were prospectively followed up. Patients were in a hospitalized cohort. Differences in outcome included in the study whether referred to hospital or not, between hospitals are more likely to reﬂect the dif- and so provided prognostic data on an unselected ferences in the proportions of patients with severe community-based cohort of patients with stroke. Studies stroke than any differences in treatment given (section of prognosis in other community-based series have pro- 17.12.7). vided broadly similar results, although higher case fatality • Selection bias: Hospital admission rates vary consider- has been reported in other populations, e.g. Belluno, ably from place to place, from country to country, and Italy. 22 However, the prognosis of patients in one’s own from time to time. For example, the admission rates in clinical practice may be different because of: community-based studies have varied between 53% • Different population characteristics: The stroke popula- in Siberia, and at least 95% in Umea, Sweden. 24,25 tion is different from that in Oxfordshire; patients These data have to be interpreted in the knowledge may be younger or older, of different racial or ethnic that deﬁnitions of hospital admission vary – e.g. did background, have more or less severe strokes, more or the patient actually spend a night in a hospital bed? less comorbidity, or a different pattern of stroke patho- And how were patients who had their stroke while in logy. For example, a greater proportion of strokes are hospital, or who were admitted late after the stroke, attributed to haemorrhage in Japan 23 (sections 5.2 and handled in the analysis? The type of hospital (e.g. dis- 18.2.2), and thus one might expect a higher early case trict general, university or tertiary referral hospital) fatality (Table 10.3). and the specialties represented in it (e.g. neurosurgery, • Referral bias: In any hospital, the prognosis of patients neurology, general medicine, care of the elderly, etc.), will be affected by referral bias (Table 10.2). In general, will have a major inﬂuence on case mix (section stroke patients referred to hospital can be expected to 17.12.7), so that clinicians working in different depart- have a worse prognosis (i.e. a higher case fatality and ments and institutions will form, from their own worse functional outcome) because a greater propor- experience, widely differing views of the prognosis of tion of milder cases are looked after by their family stroke patients. 26 doctors at home. However, this is not always predict- • Impact of time from onset: If patients are seen very early able, since in some places younger patients, who have after stroke onset (e.g. because the hospital is in a on average a better prognosis (Fig. 10.10), may be city, has an accident and emergency unit or provides a referred to hospital more often than older patients. thrombolysis service), they are likely to have a worse Furthermore, some patients with very severe strokes prognosis than patients who are seen later (such as those referred to a distant tertiary referral centre), because they must survive long enough to be admitted 100 (Table 10.2). • Follow-up method: Unless patients are followed up 80 using similar deﬁnitions of outcome and for the same Per cent surviving 60 time period as those in the published studies, their prognosis will be different (Table 10.2). Also, the reasons why patients are lost to follow-up may be 40 are lost to follow-up, this biases the prognostic data 20 related to their outcome. For example, if dead patients in a favourable direction, and if patients who have made a good recovery (and are mobile) are lost because 0 1 2 3 4 5 they move away, this biases the prognostic data in Years the opposite direction (Table 10.2). Furthermore, if all patients are not followed up, the perceived prognosis Under 65 years (n = 163) 65–74 years (n = 195) is likely to be overly inﬂuenced by the outcome of 75–84 years (n = 228) Over 84 years (n = 89) the last few patients who are remembered most vividly (i.e. recall bias), or perhaps by the patients who had Fig. 10.10 A Kaplan–Meier plot showing the proportion of patients of different ages surviving after a ﬁrst-ever-in-a- particularly good or particularly bad outcomes. lifetime stroke (reproduced with permission from Dennis • Random errors and small samples: If the estimate of the et al. 1993). 75 outcome is based on the follow-up of too few patients, .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 513 10.2 What is this patient’s prognosis? 513 it may differ from that in published studies simply by At present, it is impossible to predict an individual’s chance alone. outcome early after stroke onset with enough accuracy • Differences in management: Patients may be managed for it to be of much value in clinical practice. more or less effectively than those in Oxfordshire and other published studies, and thus the outcomes may Predicting early death be better or worse. However, the likely impact of differences in treatment between centres is likely to Clinical features such as decreased conscious level, be swamped by other factors that have a much conjugate gaze palsy, severe bilateral weakness and greater inﬂuence on outcome (e.g. case mix, section abnormal breathing patterns which – alone or in com- 17.12.7). bination – indicate severe brainstem dysfunction, due to direct damage or raised intracranial pressure, are The differences in outcome between your service and highly predictive of early death. These clinical features in those of your colleagues in other hospitals are more combination with radiological features such as massive likely to reﬂect differences in the patients you treat intracerebral haemorrhage with mass effect can help than any differences in the quality of care you guide management decisions. 27 However, many patients provide. who die do not have these features, and occasionally patients with more than one of these features make an unexpectedly good recovery. Sometimes one sees 10.2.7 Predicting outcome in individual patients patients with single predictive factors of a poor outcome, Unfortunately, it is difﬁcult to predict an individual e.g. just periodic respiration (section 11.2.2) but, in isola- patient’s outcome accurately enough for it to be of much tion, these factors can be associated with a good re- value in clinical practice. It is also difﬁcult to decide on covery; it is the combination of prognostic factors that is the acceptable accuracy of any predictive tool, because the more powerful predictor. this depends on the consequences, or cost, of getting it wrong. Taking an extreme example, if one was sure Predicting longer-term outcomes that a patient with an apparently severe stroke who was being supported on a ventilator was not going to have It is even more difﬁcult to predict longer-term outcomes. an acceptable long-term quality of life, then one might Many of the factors that predict a high early risk of death withdraw ventilatory support, particularly if the patient also predict a high risk of long-term dependency if the had left an appropriate advanced directive. However, in patient survives. In general, the patient’s age, pre-stroke this situation one would have to be very conﬁdent of health status and indicators of stroke severity (e.g. con- one’s prediction. 27 scious level, motor impairment, disability, cognitive A large number of factors are associated with better function) predict the likelihood of survival free of depend- than average or worse than average outcomes, including: ency. In predicting longer-term outcome, one also has clinical features shortly after stroke onset; the results the problem that further events – which may be related of investigations; and the patient’s progress over the to the initial stroke (e.g. recurrent strokes and myo- initial post-stroke period (Table 10.4). Many are inter- cardial infarction) or may not (e.g. development of related – e.g. conscious level and lesion size on brain unrelated illness) – can occur and have a major and often computed tomography (CT) – so multiple regression quite unpredictable effect on outcome. It is even more statistical techniques are needed to identify the factors difﬁcult to estimate an individual’s risk of further that independently predict outcome. Many studies have vascular events (section 16.2.1). focused only on factors that are associated with an increased risk of death. However, a growing number also Methods of prediction consider those factors that are related to a good or bad functional outcome, and in general, and not surpris- A variety of different approaches have been taken in ingly, these factors are similar to those that predict death predicting outcome after stroke. (Table 10.4). Unfortunately, methodological problems have so far limited the usefulness of these studies 28–30 the single-factor approach (Table 10.5). Therefore, it is useful to have a list of crite- The simplest method is to identify a single factor, the ria against which one can judge the quality of a study presence or absence of which early after the stroke indi- reporting the development of a predictive model to help cates the likelihood that the patient will have a good or decide whether it would be ‘safe’ to use in one’s own bad outcome. The most widely used examples are age practice (Table 10.6). (Fig. 10.10), reduced level of consciousness (section 11.3) .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 514 514 Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients Table 10.4 Factors that have been shown Poor survival Poor survival or poor functional to be statistically signiﬁcant (P < 0.05) outcome independent predictors of poor survival and/or poor functional outcome after Demographic features stroke in at least one study including Increasing age + + more than 100 patients. 74 Male sex + Social factors Smoking + Excess alcohol + + Unmarried + Living alone + Past history Previous stroke/TIA + + Ischaemic heart disease + Peripheral vascular disease + Diabetes mellitus + Cancer + General clinical features Lost consciousness at onset + + Post-stroke seizures + Atrial ﬁbrillation + + Cardiac failure + Fever + Pneumonia + Obstructive pulmonary disease + Renal disease + Tachycardia + High systolic blood pressure + + High diastolic blood pressure + High pulse pressure + Urinary incontinence + + Disability post-stroke + + Neurological signs Reduced level of consciousness + + Severe motor deﬁcit + + Bilateral extensor plantars + Pupil abnormality + Impaired proprioception + Visuospatial dysfunction + Cognitive impairment + + Depression + Poor balance/ataxia + Unable to walk + Total anterior circulation syndrome + + Early deterioration + Simple laboratory tests High haematocrit + High blood white cell count + High erythrocyte sedimentation rate + Hyperglycaemia + High blood urea/creatinine + High plasma cholesterol + Low arterial Po 2 + Abnormal electrocardiogram + + Complex tests (CT or MR brain scan) Large stroke lesion (haematoma) + + Site of brain lesion + + Mass effect + + Intraventricular blood (in haemorrhagic stroke) + + Visible infarction + Note: many factors have been identiﬁed in just one or two studies, and many of these factors are probably interrelated (e.g. indicating common factors such as frailty, lesion size or stroke severity). .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 515 10.2 What is this patient’s prognosis? 515 Table 10.5 Methodological problems in studies of prediction Table 10.6 Criteria to judge the quality of a study reporting the of outcome after stroke. development of a predictive model for stroke patients (adapted from Counsell 1998). 74 Failure to describe adequately the group of patients in whom the work was done Is the model externally valid? i.e. is it applicable to your patients? Use of unrepresentative cohorts of patients, e.g. highly Was the model developed in a community-based cohort, selected patients in rehabilitation settings i.e. unselected cases? Many studies are retrospective, which limits the range and Were patients with transient ischaemic attack and perhaps the reliability of the baseline and outcome data subarachnoid haemorrhage separated out? Failure to adequately deﬁne the baseline variables collected Were there any major exclusion criteria? Variation in the timing after stroke of the baseline patient Were the age and sex of the patients given? assessments Were details of any treatment given? Failure to measure outcome at a relevant and uniform point Is the model internally valid? after the stroke, i.e. 6 months post-stroke rather than at Was the delay between stroke onset and inclusion given, hospital discharge and was it short? Failure to use reliable and valid measures of outcome Were less than 10% of the cohort excluded or lost? Inadequate sample size Were baseline and outcome data collected prospectively? Failure to use appropriate statistical techniques to adjust for Was outcome measured using valid and reliable the interactions between baseline variables instruments? Failure to test the accuracy of any predictive model in an Was outcome measured at a ﬁxed point after the stroke? independent data set Was follow-up sufﬁciently long to provide useful data? Were important predictors included, e.g. stroke severity and age? Could the predictive variables be collected reliably? and urinary incontinence, which have all been related to Were statistical analyses appropriate? a poor survival and functional outcome. 30–32 Of course, Was a stepwise analysis performed? if urinary catheterization, with all its risks, becomes the Were the strengths of correlations between predictive norm, urinary incontinence loses its predictive value variables assessed (i.e. collinearity). Strong correlations (section 11.14). Measures of cognitive function at initial between predictive factors can cause multiple assessment (section 11.29) have also been related to regression techniques to give spurious results. poor functional outcomes. 30 Although such models are Was the outcome event per predictive factor ratio greater simple to use, and can guide clinical management, the than 10? user must be aware of their inaccuracy. They have a more Was the ability of the model to discriminate between obvious use in stratifying patients who are being ran- patients with a good and bad outcome tested? This is domized in clinical trials. best done by establishing the sensitivity and speciﬁcity of the model over a full range of probabilities, plotting a Reduced conscious level and urinary incontinence in receiver operating curve and establishing the area under the ﬁrst few days after stroke are both associated, in that curve (Fig. 10.11). general, with a poor outcome. Unfortunately, this Was the model calibrated to establish any bias of association is not reliable enough to be anything other predictions in grouped data? This can be done by than a very rough guide to prognosis in managing plotting the proportions predicted to have a certain individual patients. outcome against the proportion of patients who actually had that outcome in an independent test data set. Perfect calibration is indicated by the diagonal the extent of brain damage (Fig. 10.12). In general, the greater the extent, the worse the clinical Has the model been validated? outcome, except for critically sited strokes, particularly Has the model been tested in the population from which in the brainstem, where even quite small lesions can be it was derived? fatal. Many of the clinical indicators of poor prognosis Has the model been tested in an independent population? relate quite closely to the size of the brain lesion. For Has the model been compared with other predictive example, the Oxfordshire Community Stroke Project systems, including informal clinical judgement? classiﬁcation (section 4.3.8) reﬂects the extent of brain Has the model been evaluated in a randomized controlled damage, and so the prognosis of the different groups trial, i.e. is its use associated with improved outcomes? varies (Fig. 10.7). Imaging techniques, including CT, Is the model practical? single-photon emission CT, and magnetic resonance Could the predictive variables be collected in practice? Was the actual model published? imaging have so far added little to the accuracy of Were conﬁdence intervals for the model given? clinical predictors. 33–34 None of this is at all surprising, .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 516 516 Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients because the extent of the stroke lesion on brain imaging made more reliable and ‘user-friendly’, so that they do can so often be predicted from the clinical ﬁndings not require the clinician to perform complex calcula- (section 4.3.7). Moreover, a substantial proportion of tions. 41 Nonetheless, such models have been used to patients have a normal or near normal CT scan early stratify groups of patients by predicted prognosis in large after even a major ischaemic stroke, which weakens the randomized trials, in which complex calculations are early predictive utility of CT imaging (section 5.4.2). easily performed by computer during the randomiza- Other technologies, such as transcranial magnetic stimu- tion process. 42 Indeed, large randomized trials provide lation, have been shown to predict outcome though it an excellent opportunity to test such predictive models is unclear how much they add to prediction based on prospectively. 43 simpler methods. 35 stroke scales In general, the larger the stroke lesion, the worse the These provide scores which, depending on the presence, likely outcome, except for small, critically sited lesions, absence, or severity of various neurological impair- which may be associated with a poor outcome. ments, are frequently used to describe the severity of stroke in the acute phase and have also been used to pre- dict outcome. However, because strokes scales such as mathematical models These are based on regression analyses and have been the National Institutes of Health Stroke Scale (NIHSS) developed by several groups to predict both survival and reﬂect stroke severity but not other predictive factors, functional outcome. 36–37 Although these are generally a such as age or pre-stroke function, the addition of these little more accurate than models based on a single vari- latter factors is bound to add to the scale’s predictive able, any advantage may be offset by the practical difﬁ- accuracy for death or dependency. 44,45 Although the culties in applying them. 38 Also, only a couple, a simple NIHSS has similar predictive accuracy to mathematical six-variable model (Table 10.7) and the Orpington models (e.g. Guy’s and Orpington prognostic scales), it Prognostic Score have been externally validated in inde- contains items which add little to its performance (i.e. pendent cohorts of sufﬁcient size 37,39,40 (Figs 10.11, there is a lot of redundancy). 39,46,47 10.12). As one would predict, where they are externally validated they generally perform less well than in the predictions based on measures of function cohort from which they were developed (internal valida- early after a stroke tion). 31,37 If such models are to be used in routine clinical Scales such as the Barthel index have been used to pre- practice, they need to be further reﬁned, tested pro- dict eventual functional outcome and may be particularly spectively in large independent cohorts of patients, and useful for those working in rehabilitation facilities. 30 Table 10.7 A mathematical model to Variable† Parameter Odds ratio predict the probability of survival free coefﬁcient, (95% CI) from dependency in activities of daily b (SE) living 1 year after a stroke. This was derived from the Oxfordshire Community Constant 15.586 (1.748) Stroke Project using logistic regression, Age –0.085 (0.014) 0.92‡ (0.89–0.94) and has been externally validated on two Living alone 0.384 (0.259) 0.68 (0.41–1.14) independent cohorts of stroke patients Independent pre-stroke –3.174 (0.639) 25.00 (6.67–100) (Figs 10.11 and 10.12). Calculating the Normal Glasgow Coma Scale verbal –2.177 (0.504) 9.09 (3.33–25) probability of an outcome is complex (see Able to lift arms –2.319 (0.513) 10.00 (3.70–25) equation), but easily achieved using a Able to walk –1.154 (0.402) 3.12 (1.45–7.14) programmable calculator or a nomogram. Health warning – this model should not † Dichotomous variables were coded 1 = yes, 2 = no. be used to make ﬁrm predictions in ‡ Per year of age. individual patients! Y Y Probability of outcome is calculated by: P = e /(1 +e ), where Y = a+b X +b X 2 2 1 1 + ... +b X . i i Y = 15.586 – (0.085 × age) + (0.384 × living alone) – (3.174 × independent pre- stroke) – (2.177 × normal GCS verbal) – (2.319 × able to lift arms) – (1.154 × able to walk), where dichotomous variables have numeric values of 1 or 2 as described above. .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 517 10.2 What is this patient’s prognosis? 517 1.00 0.90 0.80 0.70 Sensitivity 0.60 0.50 0.40 0.30 Line representing prediction no better than chance 0.20 0.10 0.00 0.20 0.40 0.60 0.80 1.00 1–specificity (a) Community-based cohort 1.00 0.90 Fig. 10.11 Receiver operating characteristic (ROC) curves for a statistical 0.80 model (Table 10.7) to predict the 0.70 probability of survival free of dependency 1 year after stroke. The model was derived 0.60 from the Oxfordshire Community Stroke Project, and has been tested on two Sensitivity 0.50 independent cohorts of stroke patients: 0.40 (a) a community-based cohort derived Line representing prediction from the SEPIVAC and Perth Stroke 0.30 no better than chance Registries and (b) a hospital-based cohort, 0.20 the Lothian Stroke Registry. The area under the curves indicates the accuracy 0.10 of the model – i.e. the larger the area, the more discriminatory the model. The 0.00 0.20 0.40 0.60 0.80 1.00 diagonal line indicates prediction no 1–specificity better than chance. (b) Hospital-based cohort but this is bound to depend on the experience of the rate of change Some repeated measure of the patient’s condition during clinician. 50 the early clinical course can be used to predict the likely longer-term outcome. 48,49 This might be likened to the In practice, the predictive systems developed so far are growth curves used by paediatricians. Predictions for not sufﬁciently accurate to inﬂuence important clinical individuals would then depend on the pattern of recov- decisions in individuals. However, they may be useful as ery observed in large cohorts of patients. tools to: • guide less experienced clinicians in what to say to patients and carers; informal judgement The most common method of prediction is the • choose who to randomize in trials of acute treatment; informal judgements we make about patients during • help decide which patient is likely to require an our daily work. The accuracy seems to be similar to those extended period of rehabilitation; of mathematical models, at least in predicting a simple • assess the quality of care, for example by adjusting dichotomous outcome (dependent or independent), outcome data from different groups of patients for .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 518 518 Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients (a) Community-based cohort differences in case mix at baseline, one can begin to compare the quality of care given by different hospi- 1 currently available will be replaced by more precise, tals or units (section 17.12.7). Observed proportion with good outcome 0.7 more robust and better-validated models. These might 0.9 In the future, we hope the fairly crude predictive models 0.8 be used to predict not only survival and basic functional 0.6 outcome, but also the rate of recovery of individual im- pairments, disabilities and handicaps, which is so import- 0.5 ant in planning treatment. One might come to combine 0.4 several approaches to predicting a patient’s outcome, so that one relies on the mathematical modelling in the 0.3 early stages, but as more data become available from 0.2 continued observation of the patient, a patient’s speciﬁc 48 ‘recovery curve’ might be plotted. One could imagine 0.1 that these predictive systems could be presented in 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 several forms for clinical use: wall charts; clinical slide rules; on pocket calculators; palm or desktop computers, Predicted proportion with good outcome or hospital-wide computer networks. Eventually, any such system will inevitably consume resources and there- (b) Hospital-based cohort fore will have to be tested to show that it improves the 1 effectiveness and efﬁciency of stroke services. Observed proportion with good outcome 0.7 10.3 Delivering an integrated 0.9 0.8 0.6 management plan 0.5 0.4 10.3.1 Introduction 0.3 The term ‘stroke’ embraces a very wide spectrum of 0.2 clinical presentations ranging from neurological deﬁcits lasting just one day to those leading rapidly to death 0.1 or causing lifelong disability. These deﬁcits are layered 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 on a complex mix of pre-existing disease, personality, Predicted proportion with good outcome social and environmental factors. We have already seen the range of possible pathological types (Chapter 5) and Fig. 10.12 Calibration of a statistical model (Table 10.7) to causes (Chapters 6–9), but each patient requires a man- predict the probability of survival free of dependency 1 year agement plan tailored to his or her own individual needs. after a stroke (i.e. good outcome). The model was derived from the Oxfordshire Community Stroke Project, and has been For a patient whose symptoms resolve completely within tested on two independent cohorts of stroke patients: (a) a a few days, the emphasis should be on diagnosis and sec- community-based cohort derived from the SEPIVAC and Perth ondary prevention. In a patient with a major disabling Stroke Registries and (b) a hospital based cohort, the Lothian stroke, the emphasis must be on treatment of the acute Stroke Registry. The dotted line indicates perfect calibration. phase, prevention of complications and rehabilitation, The vertical lines indicate the 95% conﬁdence intervals, which although of course secondary prevention is still relevant. depend on the number of patients in each at-risk group. The As ever, the essential ﬁrst step in formulating a manage- model calibrates better in the hospital-based than in the ment plan is a full and detailed assessment. However, community-based cohort, and tends to be over-optimistic this will in practice often be staged so that one might in those with the greatest probability of a good outcome perform a rapid, less detailed assessment very early to (survival free of dependency). guide hyperacute treatment and follow this up with more detailed, time-consuming assessments later. .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 519 10.3 Delivering an integrated management plan 519 Stroke (Face Arm Speech Test) Speech impairment YES NO ? Facial palsy YES NO ? Affected side L R Arm weakness YES NO ? Affected side L R Speech If the patient attempts a conversation – Look for NEW disturbance of speech – Check with companion – Look for slurred speech – Look for word-ﬁnding difﬁculties. This can be conﬁrmed by asking the patient to name commonplace objects that may be nearby, such as a cup, chair, table, keys, pen – If there is a severe visual disturbance, place an object in the patient’s hand and ask him/her to name it Facial movements Ask patient to smile or show teeth – Look for NEW lack of symmetry – Tick the ‘YES’ box if there is an unequal smile or grimace, or obvious facial asymmetry – Note which side does not move well, and mark on the form whether it is the patient’s right or left side Arm movements – Lift the patient’s arms together to 90° if sitting, 45° if supine and ask them to hold the position for 5 seconds then let go Fig. 10.13 Face Arm Speech Test (FAST) – Does one arm drift down or fall rapidly? to judge whether a stroke is likely. – If one arm drifts down or falls, note whether it is the patient’s left or right arm ?, uncertain. the patient’s home, by a paramedic in the ambulance, or 10.3.2 Assessment a triage nurse or a trainee hospital doctor in the emer- The assessment should aim to answer the following gency department. None may have had much training questions: or experience in the assessment of patients with stroke. • is it a vascular event (transient ischaemic attack or An average general practitioner with 2000 patients stroke)? (Chapter 3); is likely to see only about four new cases of stroke • where is the lesion? (Chapters 3 and 4); each year. Simple assessment tools, such as the FAST 51,52 • what sort of stroke lesion is it (ischaemic or haemor- (Fig. 10.13) or ABCD 53,54 (Fig. 10.14), may enable non- rhagic)? (Chapter 5); experts to appropriately triage patients to the correct • what is the likely cause? (Chapters 6–9); service in the appropriate time scale to allow effective • what is this patient’s prognosis? (sections 10.2 and 16.2); treatment to be given. For example, the FAST test has • what are this patient’s particular problems? (Chap- been used to increase the proportion of patients treated ter 11). with thrombolysis, and it has been suggested that the The answers to these questions will determine the ABCD score could be used to triage transient ischaemic management of the patient and which therapeutic attack patients to ensure that those at highest risk of interventions are appropriate. The ﬁrst ﬁve questions early stroke get earliest access to secondary preven- have been discussed in previous sections, and before pro- tion. Alternatively, assessments by a non-specialist are ceeding to discuss the assessment of speciﬁc problems to increasingly being augmented by a remote specialist via complete the diagnostic formulation (Chapter 11), some telemedicine links (section 17.5.6). general principles and the organization of assessment Guidelines, a protocol or, in hospital, a clerking or must be considered. admission form may be useful tools to ensure that each In practice, the patient’s ﬁrst assessment is often patient has a thorough and relevant assessment. We, and carried out by either a general (family) practitioner in others, have demonstrated a signiﬁcant improvement .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 520 520 Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients Factor Score Age <60 years 0 >60 years 1 Blood pressure SBP <140 mmHg and DBP <90 mmHg 0 SBP >140 mmHg and/or DBP >90 mmHg 1 Clinical features No weakness or speech problem 0 Dysarthria and/or dysphasia without weakness 1 Definite weakness of face, arm or leg 2 Duration of longest TIA <10 mins 0 in last month 10–59 mins 1 ≥60 min 2 Total score Risk of stroke Approx % risk in 7 days ≤1 Low 2 Low Fig. 10.14 ABCD scale to estimate the risk of 3 Low stroke early after a transient ischaemic attack. 4 Quite low >1% 5 High >10% DBP, diastolic blood pressure; SBP, systolic blood 6 Very high >20% pressure. in the completeness of the assessment of stroke pa- swallow safely are all relevant to their care from the mo- tients following the introduction of a stroke clerking ment of hospital admission, or indeed from the moment or admission form. 55,56 The use of the form also makes of assessment at home. The components of the initial it much easier to access the information subsequently assessment and their use are summarized in Table 10.8. and to identify relevant items that are missing. However, It is important to emphasize again that assessment may one has to acknowledge the practical difﬁculties of well be staged; it is not a ‘once only’ process, but should introducing disease-speciﬁc assessment forms in situa- continue throughout the course of the patient’s recovery tions – e.g. accident and emergency departments – where from stroke. patients with a wide range of medical problems are being assessed. For example, it may not be clear until the end Assessment is not a ‘once only’ process, but should of the assessment what the most likely diagnosis is and continue throughout the course of the patient’s therefore which form should have been used. One solu- recovery from stroke. tion might be a ‘core form’ for all patients, with supple- mentary sheets for speciﬁc common conditions. There Health professionals involved in the initial assessment is reasonable evidence that patient-speciﬁc reminders of stroke patients learn that, although the patient may (e.g. ‘if ischaemic stroke, prescribe aspirin’), which can be a useful source of information, other people often be included in such forms (perhaps as part of an inte- provide more information that is essential to planning grated care pathway), improve adherence to manage- treatment. This is particularly important when the ment guidelines although randomized trials have not yet patient, for a variety of reasons, cannot communicate. demonstrated improvements in patient outcomes. 57,58 It is usually valuable to spend a little time interviewing the family, neighbours, general practitioner, ambulance A clerking or admission form will improve the technicians or nursing staff, using a telephone if neces- completeness and relevance of the initial assessment sary (Fig. 10.15). and will facilitate communication and audit, but the introduction of diagnosis-speciﬁc forms is not without To complete an assessment it is usually valuable to its problems. talk to family, neighbours, or family doctor; essential information can often be collected by a telephone call Although a physician is often the ﬁrst person to assess to the appropriate person. the patient, it is important to emphasize that assessment should often involve other members of the multidisci- Social environment plinary team (section 10.3.5). It is often very valuable, even on the day of the stroke, to involve the nurse, the The patient’s social environment is a very important physiotherapist and a speech and language therapist. factor in determining the overall effect that a stroke will Advice on the patient’s risk of pressure sores, lifting, have on an individual and his or her family. Accurate handling and positioning and the patient’s ability to knowledge of social networks is therefore critical when .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 521 10.3 Delivering an integrated management plan 521 Table 10.8 Information that should be collected during the initial assessment of a stroke patient, and the potential value. Type of Diagnosis Cause Prognosis Hyper-acute Problems Secondary information treatment prevention Demographics + + + + + History of onset + + + + Risk factors + + + + + Coexisting disease + + + + + + Medication + + + + Social details + + + + Pre-stroke function + + + General examination + + + + + + Neurological examination + + + + Investigations + + + + + + because one of the major aims of rehabilitation is to minimize handicap. For example, it may be more appro- priate to put greater energy and resources into occupa- tional therapy for a craftsman than for a schoolteacher, who might require relatively more speech therapy. The patient should, if possible, have a major role in deciding such priorities. It is so often difﬁcult to see the real person behind the facial weakness, severe aphasia, hemiplegia and incontinence. Although a lot of this background information can be collected over a longer period while the patient is recov- ering, it can be useful early on and may be more easily collected at the initial assessment. So often, the family – who may be the only source of this sort of information – disappear within a few hours of the admission and may then not reappear to be asked the relevant questions. It is Fig. 10.15 The telephone is a valuable diagnostic tool. therefore vital to seize the opportunity when a patient is ﬁrst admitted, or at least during the next day or two, to obtain as much information from the family and friends as possible. Clearly, this may not be regarded as a medical setting longer-term goals for rehabilitation and in plan- priority, but professionals other than doctors, in par- ning discharge from hospital. Moreover, this allows one ticular the nursing staff, are often well placed to collect to build up a picture of the patient as a person rather it. However, often this is not done very well. A complete than as ‘just another stroke’ or ‘that ﬁbrillating hemi- picture of the patient’s pre-stroke life will be useful plegic in bed six’. It is often difﬁcult to see the real person when deciding how aggressively to manage a patient behind the facial weakness, severe aphasia, hemiplegia (e.g. neurosurgery for obstructive hydrocephalus). Because and incontinence. Finding out about the patient’s pre- this background information is so important, but may stroke life might encourage members of the team who are not be available at the initial assessment, one needs to caring for the patient to treat him or her with more have some method of identifying which items of data are understanding and sympathy. Also, it is this background missing so they can be sought later on. The clerking or information that allows one to judge the likely effect admission form, or a patient record that is shared by the that the individual’s disabilities will have on their role different professions involved (the so-called ‘combined’ in society, i.e. the likely handicap. This is important, or ‘single-patient’ record), has the potential if properly .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 522 522 Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients used to fulﬁl this role. Many stroke units have introduced determine the patient’s management. Problems can ‘integrated care pathways’, which usually include an occur at every level of the patient’s illness – i.e. patho- admission form, multidisciplinary records and guidance logy, impairment, disability and handicap (section 10.1). on how to manage common problems (section 17.14). For example, a problem at the pathology level might be ‘diagnostic uncertainty’ or ‘raised erythrocyte sedi- It is just as important to know the home and social mentation rate’, which should lead on to further invest- circumstances of a stroke patient for early decision igation if appropriate; while a problem at the level of making (such as the desirability of emergency handicap might be that the patient provides the only operation) as for later rehabilitation and discharge income for a large family, who now have no money to from hospital. feed themselves. Having identiﬁed a problem, one can then formulate a plan to solve or at least alleviate that By the end of the initial assessment, which may be problem. Thus, a problem list can be turned into an punctuated by giving a hyperacute treatment, one action plan for the individual patient. Some problems can should have collected enough information to produce a be dealt with very simply (e.g. antibiotics for a urinary diagnostic formulation, including certainty of stroke tract infection). These are the ‘short-loop problems’ diagnosis, site and size of the brain lesion, and the likely (section 10.1.2). The goal here is simply to remove the causes. This leads on to choices about the amount of problem. Other problems such as immobility – which are further investigation required (e.g. echocardiography more complex, respond more slowly to therapy and may or not?), and enables the physician to talk to the patient require several different types of intervention – are the and/or the family about the likely diagnosis, prognosis ‘long-loop problems’ (section 10.1.2); here it is useful and management. At this stage, it is valuable to con- to set a long-term goal of removing or alleviating the sider, and even list, the particular problems the patient problem, but it is also helpful to set intermediate goals has, to ensure they have all been identiﬁed and addressed. that allow one to judge whether progress is being made It may also be useful to use a checklist of the most com- towards the long-term goal. mon ones that occur at this early stage (Table 10.9). Why a goal-orientated approach has several 10.3.3 Identifying problems and setting goals advantages The patient’s assessment, both initially and subsequently, Setting goals allows forward planning and provides a should identify any major problems, and it is these that useful focus for multidisciplinary team meetings (sec- tion 10.3.7). Intermediate goals allow members of the team to coordinate their work, assuming that goals are Table 10.9 Important things to think about during the ﬁrst achieved on time, and so improve efﬁciency. For example, day after a patient has had a severe stroke. if patients are to dress the lower half of their body, they Maintenance of a clear airway must be able to stand. If the physiotherapist can estimate Treatment of co-existing or underlying disease when the patient will be able to stand independently, the Need to review the patient’s usual medication(s) occupational therapist can plan when to start working Investigation and avoidance of fever on dressing the lower half. Setting longer-term goals can, Adequacy of oxygenation for example, allow advanced planning of a pre-discharge Swallowing ability home visit and ﬁnal discharge to the community, which Hydration can reduce the patient’s length of stay in hospital by the Nutritional status number of days or weeks needed to plan a home visit or Exclusion of urinary retention to make any necessary adaptations to the patient’s home Management of urinary incontinence before discharge (e.g. stair rails). Exclusion of fractures if the patient may have fallen If realistic goals are set, they can then be used to help Prevention of: Deep venous thrombosis motivate patients, especially if they have been involved Pressure ulcers in choosing or setting the goal. Recovery from a stroke Aspiration may be very slow, so slow that the patient and even the Trauma therapists are unable to discern any progress being made Faecal impaction towards the long-term goal (e.g. to achieve independent Protection of a ﬂaccid shoulder mobility). If one sets and achieves intermediate goals Obtaining information from and giving information to the (e.g. sitting balance), progress is more easily perceived patient and family and morale maintained. The management of patients .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 523 10.3 Delivering an integrated management plan 523 Fig. 10.16 The ‘road’ to recovery after a hemiplegic stroke, showing some mobility ‘milestones’: 1, sitting balance for 1 min; 2, standing for 10 s; 3, 10 steps unaided; 4, timed 10-m walk (Smith & Baer 1999). 59 with stroke in hospital is often allowed to drift without have problems with mobility, but usually the patient direction or leadership. The clinician who is responsible will achieve certain physical milestones on the road to for the patient waves to them on a weekly ward round, in recovery 59 (Fig. 10.16). Further improvement in mobility the belief that the therapists are actively rehabilitating can be measured by recording the time it takes the the patient. If the head of the multidisciplinary team patient to walk 10 m. 60 Thus, it is a fairly simple process maintains discipline, and encourages the setting of both to set an intermediate goal, which the patient or carer intermediate and long-term goals, drift can be avoided. can understand, in terms of the level of function and the This discipline beneﬁts the patients, the team members date by which it should be achieved. and the service as a whole. The greater efﬁciency reduces If goals are to be reproducible, and therefore useful, a length of stay in hospital and allows a greater proportion standard method for assessing attainment or not needs of stroke patients to be treated in the stroke unit without to be established and used. It is more complex to set the need for extra staff or beds. goals for problems such as language or activities of daily living (ADL) than for mobility. For example, a goal for independent dressing a patient’s top half would need to Describing goals take account of important factors including the clothes Where the goal is simply the removal of a problem, such worn, fasteners involved and the amount of prompting as a urinary tract infection, it is fairly easy to describe allowed (section 11.32.5). One could use a score on it and then measure progress, e.g. relief of symptoms any one of the huge number of measures of language and sterile urine on a repeat culture. Similarly, long- function or ADL as intermediate goals, but these scores term goals are often easy to describe, but should take into are not easily understood by patients and carers, or account the patient’s need for accommodation, physical even by the professionals using them. For example, it is and emotional support, how they might ﬁll their time unlikely to mean much to a patient, or even the team and what role they play in society. Judgements about members, to aim for a Barthel index score of 12 (out of whether these long-term goals have been achieved are a maximum of 20) within the next 2 weeks. Despite relatively straightforward. For example, if the goal is to these difﬁculties, the team should attempt to identify get a patient home to live with their family, or to return problems, specify intermediate and long-term goals, to work, one does not need any complex measures of out- and introduce some meaningful measures to determine come. In some areas, it is even fairly easy to set inter- whether progress is being made towards achieving each mediate goals. For example, many patients with stroke of them. .. ..

9781405127660_4_010.qxd 10/13/07 11:04 AM Page 524 524 Chapter 10 A practical approach to the management of stroke and transient ischaemic attack patients In every stroke patient, intermediate and long-term 10 goals should be described so that progress towards them can be measured. Moreover, everyone will feel 8 Complication a sense of achievement when the goals are met. 6 Function 10.3.4 Goal setting: a diagnostic tool 4 Goals may also be useful in identifying new or previously 2 Failing to achieve goals unrecognized problems. If a patient is not achieving the goals that have been set, it may be due to a number of 456789 3 reasons, which can be divided into team factors, patient 0 1 2 10 factors and carer factors. Time (weeks) Expected Observed Team factors If the goals are too ambitious because of inaccurate dia- Fig. 10.17 ‘Failure to thrive’ after a stroke. Deviation from the expected recovery pattern might be due to any number of gnosis, inadequate assessment or uncertainty about the factors, including recurrent stroke, infections, depression, etc. prognosis, then patients may fail to achieve them and this will have a detrimental effect on the patient’s and the team’s morale. If goals are too easy, then progress may be slower than is, in fact, possible. Also, goal setting with the concept of ‘failure to thrive’ in paediatric prac- must be realistic if one is to use it to coordinate care tice (Fig. 10.17). If a patient is failing to achieve his or her (section 10.3.3). To set realistic goals in such a way that goals (or milestones), then one needs to identify the they are more often achieved than not requires an likely reason or reasons (Tables 11.5 and 11.7). understanding of the prognosis and of the likely effect- iveness of potential interventions. We have already seen When a patient is failing to achieve his or her goals (or that accurate predictions of progress and outcome are milestones) then one must identify the reason, and if difﬁcult in individual patients (section 10.2.7), but possible do something about it. informal judgements made by the team may be more accurate, because they are based on past experience and Sometimes the patient or the carer may have different observation of the patient’s progress over a period of goals from those of the team looking after the patient. time. Another reason why a patient may not be achiev- This is ‘goal mismatch’. For example, a patient who does ing a goal might be lack of appropriate treatment. Thus, not want to live alone but would prefer to live with his progress may be hampered by too little therapy, or by the or her daughter may not achieve the level of independ- wrong sort of therapy. However, since there is currently ence expected. Therefore, when setting goals ideally one so little information about the optimum amount or the should agree them with the patient and carers – although relative effectiveness of most interventions, it is difﬁcult they may be reticent about discussing such matters to sort this out. Team members will much more often openly. It is also important to involve the patient, and have to modify their therapy based on their own experi- perhaps the carer, in setting goals, to ensure that the ence rather than on evidence from properly conducted goals are really relevant to them. Most stroke patients randomized controlled trials. are retired, so that leisure activities may be particularly important to their quality of life (section 11.33.3). The Goals should be meaningful and challenging, but patient may be less interested in a goal aiming at achievable. achieving self-care in dressing than in being able to read or do the gardening. In hospital practice, where activities of daily living (ADL) abilities often determine the length Patient and carer factors of stay, too much emphasis can be placed on ADL-related New, unrecognized medical or psychological problems goals because of the pressures on the team to make (e.g. infection, recurrent stroke or depression) may not beds available for new patients and to minimize costs present overtly – especially in their early stages – but may by discharging the patient as early as possible. There develop in a less speciﬁc way, causing a patient’s progress are however important practical difﬁculties in involv- to slow, stop, or even reverse. One can draw a parallel ing patients in goal setting, including their cognitive, .. ..

Pages:

andiny.clock

Handbook of Practical Medicine Stroke

Like this book? You can publish your book online for free in a few minutes!

Create your own flipbook

TOP SEARCH

business design fashion music health life sports home marketing children

Handbook of Practical Medicine Stroke

Read the Text Version

andiny.clock

TOP SEARCH

RELATED PUBLICATIONS