9781405127660_4_012.qxd 10/16/07 10:47 AM Page 695 References 695 interaction with Bim after transient focal cerebral 107 Pearce JM. Kernohan’s notch. Eur Neurol 2006; ischemia. J Neurosci 2004; 24(36):7879–87. 55(4):230–2. 90 Driscoll M, Gerstbrein B. Dying for a cause: invertebrate 108 Sackett DL, Strauss S, Richardson WS, Rosenberg W, genetics takes on human neurodegeneration. Nat Rev Haynes RB. Evidence-Based Medicine: How To Practice and Genet 2003; 4(3):181–94. Teach EBM. 2nd edn. Edinburgh: Churchill Livingstone, 91 Xiong ZG, Zhu XM, Chu XP, Minami M, Hey J, Wei WL 2000. et al. Neuroprotection in ischemia: blocking calcium- 109 Sackett DL, Haynes RB, Guyatt G, Tugwell P. Clinical permeable acid-sensing ion channels. Cell 2004; Epidemiology. A Basic Science for Clinical Medicine. Toronto: 118(6):687–98. Little, Brown & Co., 1991. 92 Gursoy-Ozdemir Y, Qiu J, Matsuoka N, Bolay H, Bermpohl 110 Collins R, MacMahon S. Reliable assessment of the effects D, Jin H et al. Cortical spreading depression activates and of treatment on mortality and major morbidity, I: clinical upregulates MMP-9. J Clin Invest 2004; 113(10):1447–55. trials. Lancet 2001; 357(9253):373–80. 93 Otori T, Greenberg JH, Welsh FA. Cortical spreading 111 Medical Research Council. Streptomycin treatment of depression causes a long-lasting decrease in cerebral blood pulmonary tuberculosis. Br Med J 1948; 2:769–82. flow and induces tolerance to permanent focal ischemia in 112 Yoshioka A. Use of randomisation in the Medical Research rat brain. J Cereb Blood Flow Metab 2003; 23(1):43–50. Council’s clinical trial of streptomycin in pulmonary 94 Zhao BQ, Wang S, Kim HY, Storrie H, Rosen BR, Mooney tuberculosis in the 1940s. Br Med J 2005; 317:1220–3. DJ et al. Role of matrix metalloproteinases in delayed 113 Lewis SC, Warlow CP. How to spot bias and other potential cortical responses after stroke. Nat Med 2006; 12(4):441–5. problems in randomised controlled trials. J Neurol 95 Zlokovic BV. Remodeling after stroke. Nat Med 2006; Neurosurg Psychiatry 2004; 75(2):181–7. 12(4):390–1. 114 Califf RM. Simple principles of clinical trials remain 96 Perera MN, Ma HK, Arakawa S, Howells DW, Markus RM, powerful. J Am Med Assoc 2005; 293(4):489–91. Rowe CC et al. Inflammation following stroke. J Clin 115 Dyken M, White P. Evaluation of cortisone in the Neurosci 2006; 13:1–8. treatment of cerebral infarction. J Am Med Assoc 1956; 97 Dirnagl U, Iadecola C, Moskowitz MA. Pathobiology of 162:1531–4. ischaemic stroke: an integrated view. Trends Neurosci 1999; 116 Bath FJ, Owen VE, Bath PM. Quality of full and final 22(9):391–7. publications reporting acute stroke trials: a systematic 98 Del Zoppo GJ, Ginis I, Hallenbeck JM, Iadecola C, Wang X, review. Stroke 1998; 29(10):2203–10. Feuerstein GZ. Inflammation and stroke: putative role for 117 Weaver CS, Leonardi-Bee J, Bath-Hextall FJ, Bath PM. cytokines, adhesion molecules and iNOS in brain response Sample size calculations in acute stroke trials: a systematic to ischemia. Brain Pathol 2000; 10(1):95–112. review of their reporting, characteristics, and relationship 99 Feuerstein GZ, Wang X, Barone FC. Inflammatory gene with outcome. Stroke 2004; 35(5):1216–24. expression in cerebral ischemia and trauma: potential new 118 Dorman PJ, Counsell C, Sandercock P. Reports of therapeutic targets. Ann N Y Acad Sci 1997; 825:179–93. randomized trials in acute stroke, 1955 to 1995. What 100 O’Collins VE, Macleod MR, Donnan GA, Horky LL, van der proportions were commercially sponsored? Stroke 1999; Worp BH, Howells DW. 1026 experimental treatments in 30(10):1995–8. acute stroke. Ann Neurol 2006; 59(3):467–77. 119 Kidwell CS, Liebeskind DS, Starkman S, Saver JL. Trends 101 Wahlgren NG, Ahmed N. Neuroprotection in cerebral in acute ischemic stroke trials through the 20th century. ischaemia: facts and fancies: the need for new approaches. Stroke 2001; 32(6):1349–59. Cerebrovasc Dis 2004;17 (Suppl 1):153–66. 120 Lexchin J, Bero LA, Djulbegovic B, Clark O. Pharmaceutical 102 Fagan SC, Hess DC, Machado LS, Hohnadel EJ, Pollock industry sponsorship and research outcome and quality: DM, Ergul A. Tactics for vascular protection after acute systematic review. Br Med J 2003; 326(7400):1167–70. ischemic stroke. Pharmacotherapy 2005; 25(3):387–95. 121 Bekelman JE, Li Y, Gross CP. Scope and impact of financial 103 Hurtado O, Pradillo JM, Alonso-Escolano D, Lorenzo P, conflicts of interest in biomedical research: a systematic Sobrino T, Castillo J et al. Neurorepair versus review. J Am Med Assoc 2003; 289(4):454–65. neuroprotection in stroke. Cerebrovasc Dis 2006; 122 Ioannidis JP. Effect of the statistical significance of results 21 (Suppl 2):54–63. on the time to completion and publication of randomized 104 Singhal AB, Lo EH, Dalkara T, Moskowitz MA. Advances efficacy trials. [see comment]. J Am Med Assoc 1998; in stroke neuroprotection: hyperoxia and beyond. 279(4):281–6. Neuroimaging Clin N Am 2005; 15(3):697–xiii. 123 Krzyzanowska MK, Pintilie M, Tannock IF. Factors 105 Falcao AL, Reutens DC, Markus R, Koga M, Read SJ, associated with failure to publish large randomized trials Tochon-Danguy H et al. The resistance to ischemia of presented at an oncology meeting. J Am Med Assoc 2003; white and gray matter after stroke. Ann Neurol 2004; 290(4):495–501. 56(5):695–701. 124 Donnan G, Davis SM, Kaste M, International Trial 106 Stamatovic SM, Dimitrijevic OB, Keep RF, Andjelkovic AV. Subcommittee of the International Stroke Liaison Inflammation and brain edema: new insights into the role Committee ASA. Recommendations for the relationship of chemokines and their receptors. Acta Neurochir Suppl between sponsors and investigators in the design and 2006; 96:444–50. conduct of clinical stroke trials. Stroke 2003; 34(4):1041–5. .. ..
9781405127660_4_012.qxd 10/16/07 10:47 AM Page 696 696 Chapter 12 Specific treatments for acute ischaemic stroke 125 Warlow C. Over-regulation of clinical research: a threat to 140 Asplund K, Ashburner S, Cargill K, Hux M, Lees K, public health. Clin Med 2005; 5(1):33–8. Drummond M et al. Health care resource use and stroke 126 Warlow C. The Willis Lecture 2003: evaluating treatments outcome. Multinational comparisons within the GAIN for stroke patients too slowly: time to get out of second International trial. Int J Tech Assess Hlth Care 2003; gear. Stroke 2004; 35(9):2211–19. 19(2):267–77. 127 Moher D, Schulz KF, Altman DG. The CONSORT 141 Stegmayr B, Asplund K, Hulter-Asberg K, Norrving B, statement: revised recommendations for improving the Terent A, Wester PO. Large variations in medical quality of reports of parallel-group randomised trials. antithrombotic treatment of stroke patients in hospitals. Lancet 2001; 357(9263):1191–4. Results from Riks-Stroke, the Swedish National Quality 128 Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup Register. Stroke (online) 2004; 35:E199. DF. Improving the quality of reports of meta-analyses of 142 Moussouttas MM, Lichtman JH, Krumholtz HM, Cerese J, randomised controlled trials: the QUOROM statement. Brass L. The use of heparin anticoagulation in acute Quality of Reporting of Meta-analyses. Lancet 1999; ischemic stroke among academic medical centers. 354(9193):1896–900. Stroke 1999; 30:265. 129 Schulz KF, Grimes DA. Multiplicity in randomised trials II: 143 Al Sadat A, Sunbulli M, Chaturvedi S. Use of intravenous subgroup and interim analyses. Lancet 2005; heparin by North American neurologists: do the data 365(9471):1657–61. matter? Stroke 2002; 33(6):1574–7. 130 Gent M, Barnett HJ, Sackett DL, Taylor DW. A randomized 144 Lakshminarayan K, Solid CA, Collins AJ, Anderson DC, trial of aspirin and sulfinpyrazone in patients with Herzog CA. Atrial fibrillation and stroke in the general threatened stroke. Results and methodologic issues. medicare population: a 10-year perspective (1992 to 2002). Circulation 1980; 62(6 Pt 2):V97–105. Stroke 2006; 37(8):1969–74. 131 Antiplatelet Therapy Trialists’ Collaboration. Collaborative 145 Reed SD, Cramer SC, Blough DK, Meyer K, Jarvik JJ. overview of randomised trials of antiplatelet therapy. I: Treatment with tissue plasminogen activator and inpatient Prevention of death, myocardial infarction, and stroke by mortality rates for patients with ischemic stroke treated in prolonged antiplatelet therapy in various categories of community hospitals. Stroke 2001; 32:1832–40. patients. Antiplatelet Trialists’ Collaboration. [see 146 Johnston SC, Fung LH, Gillum LA, Smith WS, Brass LM, comments]. [erratum appears in Br Med J 1994; Lichtman JH et al. Utilization of intravenous tissue-type 308(6943):1540]. Br Med J 1994; 308(6921):81–106. plasminogen activator for ischemic stroke at academic 132 The National Institute of Stroke and Neurological medical centers: the influence of ethnicity [see comment]. Disorders rt-PA Stroke Study Group. Tissue plasminogen Stroke 2001; 32(5):1061–8. activator for acute ischemic stroke. N Engl J Med 1995; 147 Heuschmann P, Berger K, Misselwitz B, Hermanek P, 333(24):1581–7. Leffmann C, Adelmann M. Frequency of thrombolytic 133 Wardlaw JM, Zoppo G, Yamaguchi T, Berge E. therapy in patients with acute ischemic stroke and the risk Thrombolysis for acute ischaemic stroke. [update of of in-hospital mortality. The German Stroke Registers Cochrane Database Syst Rev 2000; (2):CD000213; PMID: Study Group. Stroke 2003; 34:1106–13. 10796329]. Cochrane Database Syst Rev 2003; (3):CD000213. 148 Katzan L, Furlan AJ, Lloyd L, Frank J, Harper D, Hinchey J 134 Alderson P, Bero LA, Eccles M, Grilli R, Grimshaw JM, et al. Use of tissue-type plasminogen activator for acute Oxman AD et al. Effective Practice and Organisation ischemic stroke. The Cleveland Area Experience. of Care Group. About the Cochrane Collaboration J Am Med Assoc 2000; 283:1151–8. (Collaborative Review Groups (CRGs)). Cochrane Library 149 Brown D, Barsan W, Lisabeth L, Gallery M, Morgenstern L. 2005;(3). Survey of emergency physicians about recombinant tissue 135 Jamtvedt G, Young JM, Kristoffersen DT, Thomson plasminogen activator for acute ischemic stroke. Ann O’Brien MA, Oxman AD. Audit and feedback: effects on Emerg Med 2005; 46(1):56–60. professional practice and health care outcomes. [update 150 Bednar MM, Gross CE. Antiplatelet therapy in acute of Cochrane Database Syst Rev 2000; (2):CD000259; PMID: cerebral ischemia. [see comment]. Stroke 1999; 10796520]. Cochrane Database Sys Rev 2003; (3):CD000259. 30(4):887–93. 136 Sandercock P, Fraser H, Thomas B, McInnes A, Dixon S. 151 van Kooten F, Ciabattoni G, Patrono C, Dippel DW, Cochrane Stroke Group 10 years on: progress to date and Koudstaal PJ. Platelet activation and lipid peroxidation future challenges. Stroke 2003; 34(10):2537–9. in patients with acute ischemic stroke. Stroke 1997; 137 Jadad AR, Cook DJ, Jones A, Klassen TP, Tugwell P, Moher 28(8):1557–63. M et al. Methodology and reports of systematic reviews 152 Antiplatelet Therapy Trialists’ Collaboration. Collaborative and meta-analyses: a comparison of Cochrane reviews overview of randomised trials of antiplatelet therapy. III: with articles published in paper-based journals. J Am Med Reduction in venous thrombosis and pulmonary Assoc 1998; 280(3):278–80. embolism by antiplatelet prophylaxis among surgical 138 Asplund K. Stroke care in the UK: is it good enough? Lancet and medical patients. Antiplatelet Trialists’ Collaboration. Neurol 2002; 1:341. Br Med J 1994; 308(6923):235–46. 139 Asplund K. Stroke in Europe: widening gap between East 153 ISIS-2 (Second International Study of Infarct Survival) and West. Cerebrovasc Dis 1996; 6:3–6. Collaborative Group. Randomised trial of intravenous .. ..
9781405127660_4_012.qxd 10/16/07 10:47 AM Page 697 References 697 streptokinase, oral aspirin, both, or neither among 17 187 of early intracranial hemorrhage following systemic cases of suspected acute myocardial infarction: ISIS-2. [see thrombolysis for acute ischemic stroke. Cerebrovasc Dis comment]. Lancet 1988; 2(8607):349–60. 2003; 16(3):183–90. 154 Sandercock P, Collins R, Counsell C, Farrell B, Peto R, 166 Intercollegiate Stroke Working Party. National Clinical Slattery J et al. The International Stroke Trial (IST): A Guidelines for Stroke. 2nd edn. Royal College of Physicians randomised trial of aspirin, subcutaneous heparin, both, of London, 2004. or neither among 19 435 patients with acute ischaemic 167 Adams HP, Adams RJ, Brott T, del Zoppo GJ, Furlan AJ, stroke. Lancet 1997; 349:1569–81. Goldstein LB et al. Guidelines for the Early Management of 155 Chen ZM, Collins R, Liu LS, Pan HC, Peto R, Xie JX. CAST: Patients With Ischemic Stroke. A Scientific Statement From Randomised placebo-controlled trial of early aspirin use in the Stroke Council of the American Stroke Association. 20 000 patients with acute ischaemic stroke. Lancet 1997; Stroke 2003; 34, 1056–83. 349:1641–9. 168 Patrono C. Aspirin as an antiplatelet drug. N Engl J Med 156 Sandercock P, Gubitz G, Foley P, Counsell C. Antiplatelet 1994; 330(18):1287–94. therapy for acute ischaemic stroke. [see comment] [update 169 Patrono C, Ciabattoni G, Davi G. Thromboxane of Cochrane Database Syst Rev 2000; (2):CD000029; PMID: biosynthesis in cardiovascular diseases. Stroke 1990; 10796284]. Cochrane Database Syst Rev 2003; 21(12 Suppl):IV130–IV133. (2):CD000029. 170 Hop JW, Rinkel GJ, Algra A, Berkelbach van der Sprenkel 157 Chen ZM, Sandercock P, Pan HC, Counsell C, Collins R, JW, van Gijn J. Randomized pilot trial of postoperative Liu LS et al. Indications for early aspirin use in acute aspirin in subarachnoid hemorrhage. Neurology 2000; ischemic stroke : a combined analysis of 40 000 randomized 54(4):872–8. patients from the chinese acute stroke trial and the 171 Hart RG, Tonarelli S, Pearce LA. Avoiding central nervous international stroke trial. Stroke 2000; 31(6):1240–9. system bleeding during antithrombotic therapy. Recent 158 Pulmonary Embolism Prevention (PEP) Trial Collaborative data and ideas. Stroke 2005; 36:1588–93. Group. Prevention of pulmonary embolism and deep vein 172 Gubitz G, Sandercock P, Counsell C. Anticoagulants for thrombosis with low dose aspirin: Pulmonary Embolism acute ischaemic stroke. [update of Cochrane Database Syst Prevention (PEP) trial. [see comment]. Lancet 2000; Rev 2000; (2):CD000024; PMID: 10796283]. Cochrane 355(9212):1295–302. Database Syst Rev 2004; (3):CD000024. 159 Keir SL, Wardlaw JM, Sandercock PA, Chen Z. 173 Sandercock P, Counsell C, Stobbs SL. Low-molecular- Antithrombotic therapy in patients with any form of weight heparins or heparinoids versus standard intracranial haemorrhage: a systematic review of the unfractionated heparin for acute ischaemic stroke. [update available controlled studies. Cerebrovasc Dis 2002; of Cochrane Database Syst Rev 2001; (4):CD000119; PMID: 14(3–4):197–206. 11687069]. Cochrane Database Syst Rev 2005; 160 Coull BM, Williams LS, Goldstein LB, Meschia JF, (2):CD000119. Heitzman D, Chaturvedi S et al. Anticoagulants and 174 Bath PM, Iddenden R, Bath FJ. Low-molecular-weight antiplatelet agents in acute ischemic stroke: report of the heparins and heparinoids in acute ischemic stroke: Joint Stroke Guideline Development Committee of the a meta-analysis of randomized controlled trials. Stroke American Academy of Neurology and the American Stroke 2000; 31(7):1770–8. Association (a division of the American Heart Association). 175 Berge E, Sandercock P. Anticoagulants versus antiplatelet Stroke 2002; 33(7):1934–42. agents for acute ischaemic stroke. [see comment]. Cochrane 161 Toyoda K, Okada Y, Minematsu K, Kamouchi M, Fujimoto Database Syst Rev 2002; (4):CD003242. S, Ibayashi S et al. Antiplatelet therapy contributes to acute 176 Chamorro A, Busse O, Obach V, Toni D, Sandercock P, deterioration of intracerebral hemorrhage. Neurology 2005; Reverter JC et al. The rapid anticoagulation prevents 65(7):1000–4. ischemic damage study in acute stroke: final results from 162 Viswanathan A, Rakich SM, Engel C, Snider R, Rosand J, the writing committee. Cerebrovasc Dis 2005; 19(6):402–4. Greenberg SM et al. Antiplatelet use after intracerebral 177 Camerlingo M, Salvi P, Belloni G, Gamba T, Cesana BM, hemorrhage. Neurology 2006; 66(2):206–9. Mamoli A. Intravenous heparin started within the first 163 Wardlaw JM, Seymour J, Cairns J, Keir S, Lewis S, 3 hours after onset of symptoms as a treatment for acute Sandercock P. Immediate computed tomography scanning nonlacunar hemispheric cerebral infarctions. Stroke 2005; of acute stroke is cost-effective and improves quality of life. 36(11):2415–20. Stroke 2004; 35(11):2477–83. 178 Diener HC, Ringelstein EB, von KR, Landgraf H, 164 MAST-I Group. Randomised controlled trial of Koppenhagen K, Harenberg J et al. Prophylaxis of streptokinase, aspirin, and combination of both in thrombotic and embolic events in acute ischemic stroke treatment of acute ischaemic stroke. Multicentre Acute with the low-molecular-weight heparin certoparin: results Stroke Trial–Italy (MAST-I) Group. Lancet 1995; of the PROTECT Trial. Stroke 2006; 37:139–44. 346(8989):1509–14. 179 Diener HC, Ringelstein EB, von Kummer R, Langohr HD, 165 Schmulling S, Rudolf J, Strotmann-Tack T, Grond M, Bewermeyer H, Landgraf H et al. Treatment of acute Schneweis S, Sobesky J et al. Acetylsalicylic acid ischemic stroke with the low-molecular-weight heparin pretreatment, concomitant heparin therapy and the risk certoparin: results of the TOPAS trial. Therapy of Patients .. ..
9781405127660_4_012.qxd 10/16/07 10:47 AM Page 698 698 Chapter 12 Specific treatments for acute ischaemic stroke With Acute Stroke (TOPAS) Investigators. Stroke 2001; 192 Lindley RI, Wardlaw J, Sandercock P, Rimdusid P, 32(1):22–9. Lewis SC, Signorini D et al. Frequency and risk factors for 180 Wong KS, Chen C, Ng PW, Tsoi TH, Li HL, Fong WC et al. spontaneous hemorrhagic transformation of cerebral Low-molecular-weight heparin compared with aspirin for infarction. J Stroke Cerebrovasc Dis 2004; 13(6):235–46. the treatment of acute ischaemic stroke in Asian patients 193 Scottish Intercollegiate Guidelines Network. Antithrombotic with large artery occlusive disease: a randomised study. Therapy. Edinburgh: Royal College of Physicians Lancet Neurol 2007; 6:407–13. Edinburgh, 1999. 181 LaMonte MP, Nash ML, Wang DZ, Woolfenden AR, 194 Toyonaga M, Hagiwara N, Irie F, Toyoda K, Fujimoto S, Schultz J, Hursting MJ et al. Argatroban anticoagulation Hitotsumatsu T et al. [Acute cervical spinal epidural in patients with acute ischemic stroke (ARGIS-1): hematoma during antithrombotic therapy: dual warnings a randomized, placebo-controlled safety study. against antithrombotic therapy]. Rinsho Shinkeigaku 2003; Stroke 2004; 35(7):1677–82. 43(5):287–90. 182 Gubitz G, Sandercock P, Counsell C. Immediate 195 Steiner T, Rosand J, Diringer M. Intracerebral hemorrhage anticoagulant therapy for acute ischaemic stroke: a associated with oral anticoagulant therapy: current systematic review of seven randomised trials directly practices and unresolved questions. Stroke 2006; comparing different doses of the same anticoagulant. 37(1):256–62. Stroke 2000; 31:308. 196 Huttner HB, Schellinger PD, Hartmann M, Kohrmann M, 183 Bath PM, Lindenstrom E, Boysen G, De Deyn P, Friis P, Juettler E, Wikner J et al. Hematoma growth and outcome Leys D et al. Tinzaparin in acute ischaemic stroke (TAIST): in treated neurocritical care patients with intracerebral a randomised aspirin-controlled trial. [see comments]. hemorrhage related to oral anticoagulant therapy: [erratum appears in Lancet 2001; 358(9289):1276]. Lancet comparison of acute treatment strategies using vitamin k, 2001; 358(9283):702–10. fresh frozen plasma, and prothrombin complex 184 Saxena R, Lewis S, Berge E, Sandercock P, Koudstaal PJ. concentrates. Stroke 2006; 37(6):1465–70. Risk of early death and recurrent stroke and effect of 197 Wijdicks EF, Schievink WI, Brown RD, Mullany CY. Early heparin in 3169 patients with acute ischemic stroke and anticoagulation in patients with prosthetic heart valves atrial fibrillation in the International Stroke Trial. Stroke and intracerebral hematoma. [comment]. Neurology 1999; 2001; 32(10):2333–7. 52(3):676–7. 185 Berge E, Abdelnoor M, Nakstad PH, Sandset PM. Low 198 Phan TG, Koh M, Wijdicks EFM. Safety of discontinuation molecular-weight heparin versus aspirin in patients with of anticoagulation in patients with intracranial acute ischaemic stroke and atrial fibrillation: a double- hemorrhage at high thromboembolic risk. Arch blind randomised study. HAEST Study Group. Heparin in Neurol 2000; 57(12):1710–13. Acute Embolic Stroke Trial. [see comments]. Lancet 2000; 199 Phan TG, Wijdicks EF, Phan TG, Wijdicks EF. Management 355(9211):1205–10. of intracranial bleeding associated with anticoagulation: 186 Randomised controlled trial of subcutaneous calcium- balancing the risk of further bleeding against heparin in acute myocardial infarction. The SCATI (Studio thromboembolism from prosthetic heart valves. sulla Calciparina nell’Angina e nella Trombosi Ventricolare J Neurol Neurosurg Psychiatry 2001; 70(6):820–1. nell’Infarto) Group. Lancet 1989; 2(8656):182–6. 200 Eckman MH, Rosand J, Knudsen KA, Singer DE, Greenberg 187 Turpie AG, Robinson JG, Doyle DJ, Mulji AS, Mishkel GJ, SM. can patients be anticoagulated after intracerebral Sealey BJ et al. Comparison of high-dose with low-dose hemorrhage?: A decision analysis. Stroke 2003; subcutaneous heparin to prevent left ventricular mural 34(7):1710–16. thrombosis in patients with acute transmural anterior 201 Butler A, Tait RC, Eckman MH, Singer DE, Rosand J, myocardial infarction. N Engl J Med 1989; 320(6):352–7. Knudsen KA et al. Restarting oral anticoagulation after 188 Collins R, MacMahon S, Flather M, Baigent C, Remvig L, intracranial hemorrhage. Response. Stroke 2004; Mortensen S et al. Clinical effects of anticoagulant therapy 35(1):5e–6. in suspected acute myocardial infarction: systematic 202 Little SH, Massel DR. Antiplatelet and anticoagulation for overview of randomised trials. [see comment]. Br Med J patients with prosthetic heart valves. Cochrane Database 1996; 313(7058):652–9. Syst Rev 2003; (4):CD003464. 189 Cairns JA, Theroux P, Lewis HD, Jr., Ezekowitz M, Meade 203 Hacke W, Donnan G, Fieschi C, Kaste M, von Kummer R, TW. Antithrombotic agents in coronary artery disease. Broderick JP et al. Association of outcome with early stroke Chest 2001; 119(Suppl 1):228S–52S. treatment: pooled analysis of ATLANTIS, ECASS, and 190 Vaitkus PT, Berlin JA, Schwartz JS, Barnathan ES. Stroke NINDS rt-PA stroke trials. Lancet 2004; 363(9411):768–74. complicating acute myocardial infarction. A meta-analysis 204 Cornu C, Boutitie F, Candelise L, Boissel JP, Donnan G, of risk modification by anticoagulation and thrombolytic Hommel M et al. Streptokinase in acute ischemic stroke: an therapy. Arch Intern Med 1992; 152(10):2020–4. individual patient data meta-analysis: The Thrombolysis 191 Scottish Intercollegiate Guidelines Network. Management in Acute Stroke Pooling Project. Stroke 2000; of Patients with Stroke. Rehabilitation, Prevention and 31(7):1555–60. Management of Complications, and Discharge Planning. 205 Haley EC, Lyden P, Johnston KC, Hemmen TM, the TNK in Edinburgh: Royal College of Physicians Edinburgh, 2002. Stroke Investigators. A pilot dose-escalation safety study of .. ..
9781405127660_4_012.qxd 10/16/07 10:47 AM Page 699 References 699 tenecteplase in acute ischemic stroke. Stroke 2005; 220 Von Kummer R. Early major ischemic changes on 36:607–12. computed tomography should preclude use of tissue 206 Hacke W. The results of the joint analysis of two phase II plasminogen activator. Stroke 2003; 34:820–1. trials on desmoteplase in acute ischemic stroke with 221 Lyden P. Early major ischemic changes on computed treatment 3 to 9 hours after stroke onset. Cerebrovasc tomography should not preclude use of tissue Dis 2005; 19 (Suppl 2):69. plasminogen activator. Stroke 2003; 33:812–22. 207 Macleod MR, Davis SM, Mitchell P, Gerraty R, Hankey G, 222 Wardlaw JM, Mielke O. Early signs of brain infarction at Fitt G et al. Results of a multicentre, randomised controlled CT: observer reliability and outcome after thrombolytic trial of intra-arterial urokinase in the treatment of acute treatment: systematic review. Radiology 2005; posterior circulation ischaemic stroke. Cerebrovasc Dis 235(2):444–53. 2005; 20:12–17. 223 Demchuk AM, Hill MD, Barber PA, Silver B, Patel SC, 208 Fibrinolytic Therapy Trialists’ Collaborative Group. Levine SR. Importance of early ischemic computed Indications for fibrinolytic therapy in suspected acute tomography changes using ASPECTS in NINDS rt-PA myocardial infarction: collaborative overview of early Stroke Study. Stroke 2005; 36(10):2110–15. mortality and major morbidity results from all randomised 224 Dzialowski I, Hill MD, Coutts SB, Demchuk AM, Kent DM, trials of more than 1000 patients. Lancet 1994; Wunderlich O et al. Extent of early ischemic changes 343:311–22. on computed tomography (CT) before thrombolysis: 209 McCarron MO, Nicoll JA. Cerebral amyloid angiopathy prognostic value of the Alberta Stroke Program Early CT and thrombolysis-related intracerebral haemorrhage. Score in ECASS II. Stroke 2006; 37(4):973–8. Lancet Neurol 2004; 3(8):484–92. 225 Schellinger PD, Fiebach JB. Perfusion-weighted 210 Kakuda W, Thijs VN, Lansberg MG, Bammer R, Wechsler L, imaging/diffusion-weighted imaging mismatch on Kemp S et al. Clinical importance of microbleeds in MRI can now be used to select patients for recombinant patients receiving IV thrombolysis. Neurology 2005; tissue plasminogen activator beyond 3 hours: pro. [see 65(8):1175–8. comment]. Stroke 2005; 36(5):1104–5. 211 Koudstaal PJ, Stibbe J, Vermeulen M. Fatal ischaemic brain 226 Zivin JA. Perfusion-weighted imaging/diffusion-weighted oedema after early thrombolysis with tissue plasminogen imaging mismatch on MRI can now be used to select activator in acute stroke. Br Med J 1988; 297(6663):1571–4. patients for recombinant tissue plasminogen activator 212 Rudolf J, Grond M, Stenzel C, Neveling M, Heiss WD. beyond 3 hours [see comment]. Stroke 2005; Incidence of space-occupying brain edema following 36(5):1105–6. systemic thrombolysis of acute supratentorial ischemia. 227 Davis SM. Echoplanar Imaging Thrombolysis Evaluation Cerebrovasc Dis 1998; 8(3):166–71. Trial. EPITHET. http://www strokecenter 213 Kwiatkowski T, Libman RB, Frankel M, Tilley B, org/trials/TrialDetail aspx?tid=420, 2005; available from: Morgenstern L, Lu M et al. Effects of tissue plasminogen http://www.astn.org.au/epithet/Epithet_home.htm activator for acute ischemic stroke at one year. N Engl 228 Dundar Y, Hill R, Dickson R, Walley T. Comparative J Med 1999; 340:1781–7. efficacy of thrombolytics in acute myocardial infarction: 214 Sandercock P, Berge E, Dennis M, Hand P, Kwan J, a systematic review. Q J Med 2003; 96(2):103–13. Lewis S et al. A systematic review of the effectiveness, 229 Walley T, Dundar Y, Hill R, Dickson R. Superiority and cost-effectiveness and barriers to the implementation of equivalence in thrombolytic drugs: an interpretation. thrombolytic and neuroprotective treatment in the NHS. Q J Med 2003; 96(2):155–60. Health Technology Assessment 2002; 6(26):1–112. 230 Mielke O, Wardlaw J, Liu M. Thrombolysis (different doses, 215 Counsell C, Dennis M. Systematic review of prognostic routes of administration and agents) for acute ischaemic models in patients with acute stroke. Cerebrovasc Dis 2001; stroke. [update of Cochrane Database Syst Rev 2000; 12:159–70. (2):CD000514; PMID: 10796381]. Cochrane Database Syst 216 Altman DG, Bland JM. Interaction revisited: the difference Rev 2004; (4):CD000514. between two estimates. Br Med J 2003; 326(7382):219. 231 Molina CA, Saver JL, Molina CA, Saver JL. Extending 217 Lagakos SW. The challenge of subgroup analyses: reporting reperfusion therapy for acute ischemic stroke: emerging without distorting. N Engl J Med 2006; 354(16):1667–9. pharmacological, mechanical, and imaging strategies. 218 O’Fallon WM, Asplund K, Goldfrank LR, Hertzberg V, Stroke 2005; 36(10):2311–20. Ingall TJ, Louis TA. Report of the t-PA Review Committee. 232 Ciccone A, Motto C, Aritzu E, Piana A, Candelise L. Risk of 2004. http://www.ninds.nih.gov/funding/ aspirin use plus thrombolysis after acute ischaemic stroke: review_committees/t-pa_review_committee/ a further MAST-I analysis. MAST-I Collaborative Group. t-pa_committee_report.pdf Multicentre Acute Stroke Trial–Italy. Lancet 1998; 219 Simon JE, Sandler DL, Pexman JH, Hill MD, Buchan AM, 352(9131):880. Calgary Stroke Program. Is intravenous recombinant tissue 233 del Zoppo GJ, Higashida R, Furlan AJ, Pessin M, Rowley plasminogen activator (rt-PA) safe for use in patients over HA, Gent M et al. PROACT: a phase II randomized trial 80 years old with acute ischaemic stroke? The Calgary of recombinant pro-urokinase by direct arterial delivery experience. [see comment]. Age Ageing 2004; in acute middle cerebral artery stroke. Stroke 1998; 33(2):143–9. 29:4–11. .. ..
9781405127660_4_012.qxd 10/16/07 10:47 AM Page 700 700 Chapter 12 Specific treatments for acute ischaemic stroke 234 Furlan AJ, Higashida R, Wechsler LR, Gent M, Rowley H, 249 Katz JM, Segal AZ, Katz JM, Segal AZ. Should thrombolysis Kase C et al. Intra-arterial prourokinase for acute ischemic be given to a stroke patient refusing therapy due to stroke. The PROACT II Study: a randomized controlled profound anosognosia? [see comment]. Neurology 2004; trial. J Am Med Assoc 1999; 282:2003–11. 63(12):2440. 235 Adams HP, Brott TG, Furlan AJ. Guidelines for 250 Worrall BB, Chen DT, Dimberg EL, Worrall BB, Chen DT, thrombolytic therapy in stroke: a supplement to the Dimberg EL. Should thrombolysis be given to a stroke guidelines for the management of patients with acute patient refusing therapy due to profound anosognosia? ischaemic stroke. A Statement for Healthcare Professionals [comment]. Neurology 2005; 65(3):500. from a Special Writing Group of the Stroke Council, 251 Kwan J, Hand P, Sandercock P. A systematic review of American Heart Association. Stroke 1996; 27:1711–18. barriers to delivery of thrombolysis for acute stroke. 236 Ingall TJ, O’Fallon WM, Goldfrank LR, Hertzberg V, Louis [see comment]. Age Ageing 2004; 33(2):116–21. TA, Christianson TJH. Findings from the reanalysis of the 252 Ciccone A, Bonito V, Italian Neurological Society’s Study NINDS Tissue Plasminogen Activator for Acute Ischemic Group for Bioethics and Palliative Care in Neurology. Stroke Treatment Trial. Stroke 2004; 35:2418–24. Thrombolysis for acute ischemic stroke: the problem of 237 Mori E, Yoneda Y, Tabuchi M, Yoshida T, Ohkawa S, consent. Neurol Sci 2001; 22(5):339–51. Ohsumi Y et al. Intravenous recombinant tissue 253 Fleck LM, Hayes OW. Ethics and consent to treat issues in plasminogen activator in acute carotid artery territory acute stroke therapy. Emerg Med Clin N Am 2002; stroke. Neurology 1992; 42:976–82. 20(3):703–15. 238 Yamaguchi T, Hayakawa T, Kiuchi H, Japanese 254 Ciccone A. Consent to thrombolysis in acute ischaemic Thrombolysis Study Group. Intravenous tissue stroke: from trial to practice. Lancet Neurol 2003; plasminogen activator ameliorates the outcome of 2:375–8. hyperacute embolic stroke. Cerebrovasc Dis 1993; 3:269–72. 255 Bateman BT, Meyers PM, Schumacher HC, Mangla S, 239 IMS study Investigators. Combined intravenous and Pile-Spellman J. Conducting stroke research with an intra-arterial recanalization for acute ischemic stroke: exception from the requirement for informed consent. the Interventional Management of Stroke Study. [see Stroke 2003; 34(5):1317–23. comment]. Stroke 2004; 35(4):904–11. 256 Demarquay G, Derex L, Nighoghossian N, Adeleine P, 240 Khatri P, Kasner SE. Ischemic strokes after cardiac Philippeau F, Honnorat J et al. Ethical issues of informed catheterization: opportune thrombolysis candidates? consent in acute stroke: analysis of the modalities of Arch Neurol 2006; 63(6):817–21. consent in 56 patients enrolled in urgent therapeutic trials. 241 Berenstein A, Siller K, Setton A, Nelson P, Levin D, Cerebrovasc Dis 2005; 19(2):65–8. Kupersmith MJ. Intra arterial urokinase for acute ischaemic 257 Kane I, Lindley R, Lewis S, Sandercock P. Impact of stroke stroke during interventional neuroradiological procedures. syndrome and stroke severity on the process of consent in Neurology 1994; 44 (Suppl 2):A356. the Third International Stroke Trial. Cerebrovasc Dis 2006; 242 Zaidat OO, Slivka AP, Mohammad Y, Graffagnino C, 21(5–6):348–52. Smith TP, Enterline DS et al. Intra-arterial thrombolytic 258 Rosenbaum JR, Bravata DM, Concato J, Brass LM, Kim N, therapy in peri-coronary angiography ischemic stroke. Fried TR et al. Informed consent for thrombolytic therapy Stroke 2005; 36(5):1089–90. for patients with acute ischemic stroke treated in routine 243 Hankey G, Warlow C. Treatment and secondary clinical practice. [see comment]. Stroke 2004; prevention of stroke: evidence, costs and effects on 35(9):e353–e355. individuals and populations. Lancet 1999; 354:1457–63. 259 Johnson DM, Kramer DC, Cohen E, Rochon M, Rosner M, 244 Gilligan AK, Thrift AG, Sturm JW, Dewey HM, Macdonell Weinberger J. Thrombolytic therapy for acute stroke in RA, Donnan GA. Stroke units, tissue plasminogen late pregnancy with intra-arterial recombinant tissue activator, aspirin and neuroprotection: which stroke plasminogen activator. Stroke 2005; 36(6):e53–e55. intervention could provide the greatest community 260 Wein TH, Hickenbottom SL, Morgenstern LB, Demchuk benefit? Cerebrovasc Dis 2005; 20(4):239–44. AM, Grotta JC. Safety of tissue plasminogen activator for 245 Anonymous. Summary of product characteristics: Actilyse. acute stroke in menstruating women. Stroke 2002; Bracknell, UK: Boehringer Ingelheim, 2003. 33(10):2506–8. 246 Weintraub MI. Thrombolysis (tissue plasminogen 261 Chalela JA, Katzan I, Liebeskind DS, Rasmussen P, Zaidat activator) in stroke: a medicolegal quagmire. Stroke 2006; O, Suarez JI et al. Safety of intra-arterial thrombolysis in 37(7):1917–22. the postoperative period. Stroke 2001; 32(6):1365–9. 247 Ciccone A, Sterzi R, Crespi V, Defanti CA, Pasetti C, 262 Kane I, Sandercock P, Thomas B. Can patients with Bioethics and Neurology Study Group of the Italian unruptured intracranial aneurysms be treated with Neurological Society. Thrombolysis for acute ischemic thrombolysis? A short systematic review. Cerebrovasc Dis stroke: the patient’s point of view. Cerebrovasc Dis 2001; 2005; 20(1):51–2. 12(4):335–40. 263 Talkad A, Mathews M, Honings D, Jahnel J, Wang D. 248 Koops L, Lindley R. Thrombolysis for acute ischaemic Reversal of warfarin-induced anticoagulation with factor stroke: consumer involvement in the design of new VIIa prior to rt-PA in acute stroke. Neurology 2005; randomised controlled trial. Br Med J 2002; 325:415–17. 64(8):1480–1. .. ..
9781405127660_4_012.qxd 10/16/07 10:47 AM Page 701 References 701 264 Linfante I, Reddy AS, Andreone V, Caplan LR, Selim M, 277 Bukow SC, Daffertshofer M, Hennerici MG, Bukow SC, Hirsch JA. Intra-arterial thrombolysis in patients treated Daffertshofer M, Hennerici MG. Tirofiban for the with warfarin. Cerebrovasc Dis 2005; 19(2):133–5. treatment of ischaemic stroke. Expert Opin Pharmacother 265 Tanne D, Kasner ES, Demchuk AM, Koren-Morag N, 2006; 7(1):73–9. Hanson S, Grond M et al. Markers of increased risk of 278 Velianou JL, Strauss BH, Kreatsoulas C, Pericak D, intracerebral hemorrhage after intravenous recombinant Natarajan MK. Evaluation of the role of abciximab tissue plasminogen activator therapy for acute ischemic (Reopro) as a rescue agent during percutaneous coronary stroke in clinical practice: the multicenter rt-PA acute interventions: in-hospital and six-month outcomes. [see stroke survey. Circulation 2002; 105:1679–85. comments]. Cathet Cardiovasc Interv 2000; 51(2):138–44. 266 Kwan J, Hand P, Sandercock P. Improving the efficiency 279 Fiorella D, Albuquerque FC, Han P, McDougall CG. of delivery of thrombolysis for acute stroke: a systematic Strategies for the management of intraprocedural review. Q J Med 2004; 97(5):273–9. thromboembolic complications with abciximab 267 Hacke W. Third European Acute Stroke Study (ECASS-III): (ReoPro). Neurosurgery 2004; 54(5):1089–97. Placebo controlled trial of alteplase (rt-PA) in acute 280 Arab D, Yahia AM, Qureshi AI. Use of intravenous ischemic hemispheric stroke where thrombolysis is abciximab as adjunctive therapy for carotid angioplasty initiated between 3 and 4 hours after stroke onset. and stent placement. Int J Cardiovasc Interv 2003; 5(2):61–6. http://www strokecenter org/trials/trialDetail 281 Chan AW, Yadav JS, Bhatt DL, Bajzer CT, Gum PA, Roffi M aspx?tid=475&search_string=ECASS 2005; available from et al. Comparison of the safety and efficacy of emboli http://www.strokecenter.org/trials/trialDetail.aspx? prevention devices versus platelet glycoprotein IIb/IIIa tid=475&search_string=ECASS. inhibition during carotid stenting. Am J Cardiol 2005; 268 Arnold M, Schroth G, Nedeltchev K, Loher T, Remonda L, 95(6):791–5. Stepper F et al. Intra-arterial thrombolysis in 100 patients 282 Velat GJ, Burry MV, Eskioglu E, Dettorre RR, Firment CS, with acute stroke due to middle cerebral artery occlusion. Mericle RA et al. The use of abciximab in the treatment Stroke 2002; 33(7):1828–33. of acute cerebral thromboembolic events during 269 Fernandes HM, Gregson B, Siddique S, Mendelow AD. neuroendovascular procedures. Surg Neurol 2006; Surgery in intracerebral hemorrhage. The uncertainty 65(4):352–8. continues. Stroke 2000; 31(10):2511–16. 283 Sherman DG, Atkinson RP, Chippendale T, Levin KA, 270 Mendelow AD, Gregson BA, Fernandes HM, Murray GD, Ng K, Futrell N et al. Intravenous ancrod for treatment Teasdale GM, Hope DT et al. Early surgery versus initial of acute ischemic stroke: the STAT study: a randomized conservative treatment in patients with spontaneous controlled trial. Stroke Treatment with Ancrod Trial. supratentorial intracerebral haematomas in the J Am Med Assoc 2000; 283(18):2395–403. International Surgical Trial in Intracerebral Haemorrhage 284 Liu M, Counsell C, Zhao XL, Wardlaw J. Fibrinogen (STICH): a randomised trial. [see comment]. Lancet 2005; depleting agents for acute ischaemic stroke. [update of 365(9457):387–97. Cochrane Database Syst Rev 2000; (2):CD000091; PMID: 271 Hill MD, Lye T, Moss H, Barber PA, Demchuk AM, 10796295]. Cochrane Database Syst Rev 2003; Newcommon NJ et al. Hemi-orolingual angioedema and (3):CD000091. ACE inhibition after alteplase treatment of stroke. 285 Hennerici MG, Kay R, Bogousslavsky J, Lenzi GL, Neurology 2003; 60(9):1525–7. Verstraete M, Orgogozo JM. Intravenous ancrod for acute 272 Engelter ST, Fluri F, Buitrago-Tellez C, Marsch S, Steck AJ, ischaemic stroke in the European Stroke Treatment with Ruegg S et al. Life-threatening orolingual angioedema Ancrod Trial: a randomised controlled trial. Lancet 2006; during thrombolysis in acute ischemic stroke. J Neurol 368(9550):1871–8. 2005; 252(10):1167–70. 286 Keeley EC, Boura JA, Grines CL. Primary angioplasty versus 273 Montalescot G, Barragan P, Wittenberg O, Ecollan P, intravenous thrombolytic therapy for acute myocardial Elhadad S, Villain P et al. Platelet glycoprotein IIb/IIIa infarction: a quantitative review of 23 randomised trials. inhibition with coronary stenting for acute myocardial [see comment]. Lancet 2003; 361(9351):13–20. infarction. N Engl J Med 2001; 344(25):1895–903. 287 Daffertshofer M, Hennerici M. Ultrasound in the treatment 274 De Luca G, Suryapranata H, Stone GW, Antoniucci D, of ischaemic stroke. Lancet Neurol 2003; 2(5):283–90. Tcheng JE, Neumann FJ et al. Abciximab as adjunctive 288 Furlan AJ, Fisher M. Devices, drugs, and the Food and Drug therapy to reperfusion in acute ST-segment elevation Administration: increasing implications for ischemic myocardial infarction: a meta-analysis of randomized stroke. Stroke 2005; 36(2):398–9. trials. J Am Med Assoc 2005; 293(14):1759–65. 289 Noser EA, Shaltoni HM, Hall CE, Alexandrov AV, Garami Z, 275 Abciximab Emergent Stroke Treatment Trial Investigators. Cacayorin ED et al. Aggressive mechanical clot disruption: Emergency administration of abciximab for treatment a safe adjunct to thrombolytic therapy in acute stroke? of patients with acute ischemic stroke: results of Stroke 2005; 36(2):292–6. a randomized phase 2 trial. Stroke 2005; 290 Smith WS, Sung G, Starkman S, Saver JL, Kidwell CS, 36(4):880–90. Gobin YP et al. Safety and efficacy of mechanical 276 Ringleb PA. Thrombolytics, anticoagulants, and embolectomy in acute ischemic stroke: Results antiplatelet agents. Stroke 2006; 37(2):312–13. of the MERCI trial. Stroke 2005; 36(7):1432–8. .. ..
9781405127660_4_012.qxd 10/16/07 10:47 AM Page 702 702 Chapter 12 Specific treatments for acute ischaemic stroke 291 Becker KJ, Brott TG. Approval of the MERCI clot retriever: 306 Fisher M, Stroke Therapy Academic Industry Roundtable. a critical view. Stroke 2005; 36(2):400–3. Recommendations for advancing development of acute 292 Kidwell CS, Jahan R, Starkman S, Saver J. MR and stroke therapies: Stroke Therapy Academic Industry recanalization of stroke clots using embolectomy (MR Roundtable 3. Stroke 2003; 34(6):1539–46. RESCUE). Stroke 2005; International Stroke Conference 307 Fisher M. Enhancing the development and approval 2005. Ongoing Clinical Trials Abstracts. of acute stroke therapies Stroke Therapy Academic 293 Tomsick TA. Mechanical embolus removal: a new day Industry Roundtable. Stroke 2005; 36:1808–13. dawning. [comment]. Stroke 2005; 36(7):1439–40. 308 Sandercock P, Roberts I. Systematic reviews of animal 294 Alexandrov AV, Molina CA, Grotta JC, Garami Z, Ford SR, experiments. Lancet 2002; 360(9333):586. varez-Sabin J et al. Ultrasound-enhanced systemic 309 Pound P, Ebrahim S, Sandercock P, Bracken MB, Roberts I, thrombolysis for acute ischemic stroke. [see comment]. Reviewing Animal Trials Systematically (RATS) Group. NEJM 2004; 18; 351:2170–8. Where is the evidence that animal research benefits 295 Qizilbash N, Lewington SL, Lopez-Arrieta JM, Qizilbash N, humans? [see comment]. Br Med J 2004; Lewington SL, Lopez-Arrieta JM. Corticosteroids for acute 328(7438):514–17. ischaemic stroke. [update of Cochrane Database Syst Rev 310 Macleod MR, O’Collins T, Howells DW, Donnan G. Pooling 2000; (2):CD000064; PMID: 10796290]. Cochrane Database of animal experimental data reveals influence of study Syst Rev 2002; (2):CD000064. design and publication bias. Stroke 2004; 35(5):1203–8. 296 Righetti E, Celani MG, Cantisani T, Sterzi R, Boysen G, 311 Macleod MR, Ebrahim S, Roberts I. Surveying the literature Ricci S. Glycerol for acute stroke. [update of Cochrane from animal experiments: systematic review and meta- Database Syst Rev 2000; (4):CD000096; PMID: 11034673]. analysis are important contributions. [comment]. Br Med J Cochrane Database Syst Rev 2004; (2):CD000096. 2005; 331(7508):110. 297 Bereczki D, Liu M, Prado GF, Fekete I. Cochrane report: 312 Lees KR, Zivin JA, Ashwood T, Davalos A, Davis SM, A systematic review of mannitol therapy for acute Diener HC et al. NXY-059 for acute ischemic stroke. ischemic stroke and cerebral parenchymal hemorrhage. N Engl J Med 2006; 354(6):588–600. [see comment]. Stroke 2000; 31(11):2719–22. 313 Renovis Inc. NXY-059 does not meet efficacy endpoints in 298 Roberts I, Schierhout G, Wakai A. Mannitol for acute phase III trial for acute ischemic stroke. 2006; available traumatic brain injury. [update of Cochrane Database Syst from: http://www.medicalnewstoday.com/ Rev 2000; (2):CD001049; PMID: 10796744]. Cochrane medicalnews.php?newsid=55212. Database Syst Rev 2003; (2):CD001049. 314 Asplund K. Haemodilution for acute ischaemic stroke. 299 Georgiadis D, Schwarz S, Aschoff A, Schwab S. [update of Cochrane Database Syst Rev 2000; (2):CD000103; Hemicraniectomy and moderate hypothermia in PMID: 10796299]. Cochrane Database Syst Rev 2002; patients with severe ischemic stroke. Stroke 2002; (4):CD000103. 33(6):1584–8. 315 Horn J, Limburg M. Calcium antagonists for ischemic 300 Morley NC, Berge E, Cruz-Flores S, Whittle IR. Surgical stroke. A systematic review. Stroke 2001; 32:570–6. decompression for cerebral oedema in acute ischaemic 316 Cochrane Handbook for Systematic Reviews of Interventions stroke. Cochrane Database Syst Rev 2002; (3):CD003435. 4.2.5 [updated May 2005]. In: Higgins JPT, Green S, editors. 301 Frank J. HeaDDFIRST. Hemicraniectomy and Durotomy In: The Cochrane Library, Issue 3, 2005. Chichester, UK: for Deterioration From Infarction Relating Swelling Trial. John Wiley & Sons, Ltd; 2005. 2000; available from http://www.strokecenter.org/trials/ 317 Altman DG, Machin D, Bryant TN, Gardner M. Statistics TrialDetail.aspx?tid=70. with Confidence. 2nd ed. London: BMJ Books; 2000. 302 The DESTINY Study Group. DESTINY – DEcompressive 318 Ludlam C, Bennett B, Fox KA, Lowe G, Reid A. Guidelines Surgery for the Treatment of malignant INfarction of the for the use of thrombolytic therapy. Haemostasis and middle cerebral arterY: Preliminary Results. Cerebrovasc Dis Thrombosis Task Force of the British Committee for 2006; 21 (Suppl 4):59–60. Standards in Haematology. Blood Coag Fibrin 1995; 303 Vahedi K, Vicaut E, Blanquet A, Payen D, Bousser MG. 6(3):273–85. DECIMAL trial: DEcompressive Craniectomy In 319 Plum F, Posner JB. The Diagnosis of Stupor and Coma, MALignant Middle Cerebral Artery Infarcts. 2004; available Philadelphia, PA: F.A. Davis & Co., 1985. from http://www.strokecenter.org/trials/ 320 Sherman DG, Albers GW, Bladin C, Fieschi C, Gabbai AA, TrialDetail.aspx?tid=564. Kase CS et al. The efficacy and safety of enoxaparin versus 304 Hofmeijer J, Amelink GJ, Algra A, van GJ, Macleod MR, unfractionated heparin for the prevention of venous Kappelle LJ et al. Hemicraniectomy after middle cerebral thromboembolism after acute ischaemic stroke (PREVAIL artery infarction with life-threatening Edema trial Study): an open-label randomised comparison. Lancet (HAMLET). Protocol for a randomised controlled trial of 2007; 369:1347–55. decompressive surgery in space-occupying hemispheric 321 Vahedi K, Hofmeijer J, Juettler E, Vicaut E, George B, infarction. Trials 2006; 7:29. Algra A et al. Early decompressive surgery in malignant 305 Schierhout G, Roberts I. Hyperventilation therapy for infarction of the middle cerebral artery: a pooled analysis acute traumatic brain injury. Cochrane Database Syst Rev of three randomised controlled trials. Lancet Neurol 2007; 2000; (2):CD000566. 6:215–22. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 703 13 Specific treatment of intracerebral haemorrhage 13.1 Pathophysiology 703 13.2 Prognosis and predictive models 703 13.3 Management 704 of cerebral autoregulation in the vasculature supplying 13.1 Pathophysiology the region of the haemorrhage. Some perifocal ischaemic damage occurs at the time of bleeding and cannot be prevented, but the question to be considered here is The events following non-traumatic spontaneous intra- whether the vicious cycle of ongoing ischaemia causing cerebral haemorrhage (ICH) have been most intensively steadily increasing pressure can be interrupted in its studied for the most common type, rupture of one or early stages. more deep perforating arteries. These events are surpris- Hydrocephalus may be an additional space-occupying 6 ingly complex, to begin with because bleeding from a factor. This complication is especially likely to occur single site often leads adjacent perforators to rupture as with cerebellar haemorrhages, but a large haemorrhage well (section 8.2.1). Furthermore, the extravasated blood in the region of the basal ganglia may also cause enlarge- inevitably causes disruption of white matter tracts and ment of the ventricular system, by rupture into the third irreversible damage to neurones in the deep nuclei or ventricle, or through dilatation of the opposite lateral ven- cortex. The protective encasement of the skull may tricle, with midline shift and obstruction of the third become a disadvantage with any sudden increase in ventricle, while the ipsilateral ventricle is compressed. volume within the intracranial cavity, as with other causes of brain swelling and space-occupying lesions. Apart from the brain tissue destroyed by the haemorrhage itself, the attendant increase in intracranial pressure may threaten other parts of the brain, particularly – but not 13.2 Prognosis and predictive models exclusively – when the intracranial pressure reaches levels of the same order of magnitude as the arterial pressure, bringing the cerebral perfusion pressure close to zero. 1 According to population-based studies the short-term Direct mechanical compression of the brain tissue prognosis of patients with ICH is poor, with over 40% surrounding the haemorrhage and, to some extent, dying within the first month. 7,8 Several hospital-based vasoconstrictor and pro-inflammatory substances in studies have attempted to develop models to predict extravasated blood also lead to impaired blood supply. 2,3 prognosis based on patients’ characteristics on admis- Cellular ischaemia leads to further swelling from sion, but almost without exception these have not been oedema, 4,5 which is initially cytotoxic and later vaso- externally validated (in independent data sets). An accur- genic (section 12.1.5). The zone of ischaemia around ate prognostic model might theoretically help in treat- the haemorrhage may swell through systemic factors ment decisions, particularly in selecting which patients such as hypotension or hypoxia. Often there is also loss should benefit from intensive care, and in informing relatives about the chances of recovery. The most recent systematic review selected 18 hospital- based studies performed before 2004 that met essential Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, criteria (multivariate analysis with logistic regression; no M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published surgical treatment; paper in one of four major European 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0. languages); the models proposed in these studies were 703 ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 704 704 Chapter 13 Specific treatment of intracerebral haemorrhage then validated against the outcome in 122 other patients with intracerebral haemorrhage admitted to a single 13.3 Management 9 centre. Most prediction models proved easy to apply and most could generate a high probability of death or dependency in patients with the combination of factors 13.3.1 Initial management providing the highest probability of poor outcome (most often the level of consciousness on admission, age, Admission to a specialized stroke unit is beneficial, as volume of haemorrhage and intraventricular extension). for stroke patients in general. 21 Once the patient’s Nevertheless, among the 10 prognostic models that could ventilation and circulation have been secured and the identity patients with a greater than 90% probability diagnosis of intracerebral haemorrhage confirmed by of death or poor outcome, only a small proportion of computed tomography (CT) scanning or magnetic reson- patients actually had this combination of unfavourable ance imaging (MRI), the next step is often to identify characteristics (5–48%, median 30%). Also the accur- the cause before any specific treatment is initiated acy of the prediction of 30-day case fatality was far from (Chapter 8). For example, if surgical evacuation of a life- ideal: in these 10 subsets of the validation series the case threatening haemorrhage is considered at all, medical fatality ranged between 67 and 100% (median 89%); 9 measures may have to be taken first (such as substitu- similarly inaccurate was a later predictive model, the tion of clotting factors in patients with liver failure or so-called Essen ICH score. 10 Early withdrawal of support on anticoagulants). Also the surgeon should think twice may then result in a self-fulfilling prophecy without if amyloid angiopathy is a likely cause (section 13.3.7). scientific foundation. 11 Clearly characteristics on admis- Repeat CT scanning is indicated if there is clinical sion alone, used in the presently available prognostic deterioration, which may be caused by rebleeding at models, do not allow accurate prediction of outcome in a the same or sometimes at a distant site, or by hydro- sufficient proportion of patients to be clinically useful. cephalus; if there are no structural changes, the search Single-centre studies published after the systematic for systemic disorders should be intensified (section review cited above claimed that high levels of D-dimer 11.5). or troponin, or low levels of cholesterol or triglycerides Recombinant factor (rF) VIIa is known to decrease the in serum, are independently associated with poor out- severity of haemorrhage in certain surgical settings, also come; 12–14 obviously these findings require confirmation. in patients without clotting abnormalities. 22 A Cochrane review of four phase II randomized clinical trials involv- Early characteristics predicting the prognosis for ing adults aged 18 years or over, within 4 h of ICH survival of patients with intracerebral haemorrhage found a reduced relative risk of death or dependence are: level of consciousness; age; volume of on the modified Rankin scale within 90 days of disease haemorrhage and intraventricular extension of onset; 0.79, 95% confidence interval (CI) 0.67–0.93 haemorrhage, but so far predictive models apply (Fig. 13.1), but not on the extended Glasgow Outcome to only a small proportion of patients and are not Scale; 0.90 (95% CI, 0.81–1.01). 23 There was a statistic- sufficiently accurate for clinical decision making. ally significant excess of arterial thromboembolism at 160 µg/kg rFVIIa. In conclusion, adults with acute ICH One way to improve models to predict poor outcome may benefit from haemostatic therapy with rFVIIa, is to restrict the analysis to specific subgroups of ICH, but the evidence on major clinical outcomes is neither according to location or presumed cause. This has been robust nor precise. Claims that this treatment is cost- attempted for lobar haemorrhages, 15 haemorrhages in effective are based on the point estimate of benefit and the caudate nucleus 16 and pontine haemorrhages, 17 but are therefore premature. 24 again there has been no external validation. The blood pressure is usually increased in patients with Perhaps a more fruitful approach is to move away from ICH, through pre-existing hypertension, a response to a characteristics on first admission and to base predictive sudden increase in intracranial pressure, or both these models more on the clinical course in the first 24 h, 9 factors. 25 There are theoretical arguments in favour of if only because expansion of the haemorrhage occurs decreasing the blood pressure (in the hope of stopping in a substantial proportion of patients within that ongoing bleeding from ruptured small arteries), 26 as period (section 8.2.1); not surprisingly this is an adverse well as in favour of increasing it further (in the hope factor. 18 Additional determinants in the early clinical of salvaging marginally perfused areas of brain that course that probably should be included in the equation are compressed around the haemorrhage). In the are fever 19 and the development of oedema around the absence of even the flimsiest evidence supporting one haemorrhage. 20 or other point of view (not surprisingly, a randomized .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 705 13.3 Management 705 Study rFVIla Placebo Relative risk (fixed) Weight Relative risk (fixed) n/N n/N 95% CI (%) 95% CI rFVIla EurAsia llA 16/36 5/11 6.7 0.98 [0.46, 2.06] rFVIla llB 160/303 66/96 87.7 0.77 [0.65, 0.91] rFVIla USA llA 15/32 4/8 5.6 0.94 [0.43, 2.06] Total (95% CI) 371 115 100.0 0.79 [0.67, 0.93] Total events: 191 (rFVIla). 75 (Placebo) 2 Test for heterogeneity chi-square = 0.61 df = 2 P = 0.74 I = 0.0% Test for overall, effect Z = 2.77 P = 0.006 0.1 0.2 0.5 1 2510 Favours rFVIla Favours placebo Fig. 13.1 Meta-analysis of recombinant factor VIIa for to the amount of data in the trial. The diamond represents the intracerebral haemorrhage. Each trial is represented by a overall treatment effect, and its width the 95% confidence horizontal line, the length of which describes the 95% interval. The outcome is death or dependency (modified confidence interval around the point estimate of the treatment Rankin scale 4–6) at day 90. 23 effect represented by a square, the size of which is proportional trial with 50 patients found no difference in outcome), 27 Cerebrospinal fluid drainage the only rational course is to leave the blood pressure alone in the acute phase unless it is so high that organ Insertion of a catheter in the lateral ventricle may be a damage develops (especially in the retina and kidney). definitive treatment in patients with cerebellar haemor- This is essentially the same for patients with ischaemic rhage and no signs of direct compression of the brain- stroke (section 11.7.1). This is not to say that the patient stem (section 13.3.6). But, in patients with supratentorial should be discharged with a high blood pressure; a single haemorrhage and hydrocephalus, the benefits of CSF observational study supports the notion that the long- diversion are decidedly uncertain. In a series of 22 pa- term risk not only of vascular events in general but also tients with supratentorial ICH and hydrocephalus who of rebleeding may be lowered by adequate long-term were treated with ventriculostomy, intracranial pressure blood pressure control. 28 was controlled at < 20 mmHg in 20, but only a single patient showed any improvement in not only the degree of hydrocephalus but also in the level of conscious- 13.3.2 Reduction of intracranial pressure 32 ness. Just three patients, with small haemorrhage vol- Apart from the local effects of the haemorrhage itself umes, survived to 3 months. An alternative strategy is and the subsequent perifocal oedema, other factors also to divert the CSF internally, through callosotomy and contribute to raised intracranial pressure. Frequently, fenestration of the septum pellucidum. 33 Controlled these can be treated more directly than the cerebral studies for all these procedures are woefully lacking. lesion itself, so it is important to be aware of them. They In patients with extensive intraventricular haemor- include fever, hypoxia, hypertension, seizures and eleva- rhage secondary to deep intraparenchymal or aneurysmal tions of intrathoracic pressure. A vexing and unsolved rupture, drainage of CSF is also performed in some question is whether it is useful to measure the intra- centres, in an attempt to improve the dismal prognosis. cranial pressure by the introduction of an intraventricular Systematic reviews of observational studies suggest catheter or other device, at least in patients with a that this procedure may be helpful, especially when decreased level of consciousness. Such an approach can it is combined with instillation of fibrinolytic drugs, certainly be rationalized. 29 On the other hand, any inva- but only direct comparisons in a randomized trial can sive procedure may cause complications, 30 while the pre- definitively establish the value of these therapeutic sumed advantages in terms of outcome have not been measures. 34,35 Later observational studies have not subjected to controlled studies. For what it is worth, solved the problem. 36,37 an observational comparison in traumatic brain injury failed to find any advantage of management based on Hyperventilation intracranial pressure monitoring. 31 We prefer to err on the side of caution and base management on clinical and There is no doubt that hyperventilation decreases radiological features rather than the ‘pressure’ as meas- intracranial pressure, but at the same time there is no ured directly. controlled evidence to allay concerns that the cure may .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 706 706 Chapter 13 Specific treatment of intracerebral haemorrhage be worse than the disease. The intracranial pressure goes questionable for patients with spontaneous intracerebral down because hypocapnia (usually down to values of the haemorrhages who are managed conservatively. With order of 4 kPa) causes vasoconstriction; in other words, mannitol infusions, at the very least the central venous ischaemia by compression is exchanged for ischaemia by pressure should be monitored, and kept between 5 and vasoconstriction. 38 In head-injured patients, a single and 12 mmHg, to prevent hypovolaemia. so far the only randomized controlled trial of prolonged For intravenous glycerol, meta-analysis of one minute hyperventilation not only failed to show any benefit trial (eight patients) and one moderately-sized controlled but also raised concerns about potential harm. 39,40 For trial in patients with intracerebral haemorrhage failed to patients with ICH, no controlled trials of hyperventil- find evidence of benefit. 43,48 ation have been done, but the experience with head Hypertonic saline is apparently ‘emerging’ in neuro- trauma is not encouraging. Provisionally it seems best surgical practice as an alternative to other agents, 49 but to reserve this treatment for bridging a few hours in no controlled studies are available. patients for whom surgical evacuation is planned (sec- tions 13.3.5 and 13.3.6). Corticosteroids Corticosteroids reduce peritumoral oedema and at first Osmotic agents sight it seems rational to expect a similar effect on Osmotic agents include mannitol, glycerol and sorbitol; oedematous brain tissue around an intracerebral hae- they are administered by intravenous infusion. 41 Iodide- morrhage. To date there have been four relevant but containing contrast agents used in CT scanning or angio- small trials, which have been summed up in a Cochrane graphy may inadvertently have the same dehydrating review. 50 At 1 month after randomization 57 of 92 pa- effect. 42 Osmotic agents extract water from the extra- tients (62%) allocated to glucocorticoid treatment had cellular into the intravascular compartment of the brain. died, compared with 50 of 94 patients (53%) allocated to Whether they dehydrate mainly normal instead of the control group; relative risk 1.14 (95% CI, 0.91–1.42). damaged brain tissue is a matter of debate. 41 A disturbed Treatment with glucocorticoids did not reduce poor blood–brain barrier may facilitate entry of hyperosmotic outcome at 1 month after randomization; relative agents into brain tissue, thereby decreasing the osmotic risk, 0.95 (95% CI, 0.83–1.09; Fig. 13.2). The risk of any gradient needed for effective dehydration, but on the adverse effects in the treatment group (13%) was higher other hand it may facilitate movement of fluid from than that in the control group patients (8.8%) but the brain tissue to blood. The interval between the infusion confidence interval was very wide; relative risk, RR 1.48 and the decrease of intracranial pressure is usually less (95% CI, 0.87–2.51). The most common adverse effects than 1 h, but the effect of a single dose lasts no more were infections (23%), exacerbation of diabetes mellitus than 4–6 h because the concentration of the solute (11%) and gastrointestinal bleeding (5%) but none of becomes equilibrated between the intravascular and them reached a statistically significant difference between extracellular compartments. Potential dangers of this the treatment and control groups. Thus, any benefit treatment include hypotension, hypokalaemia, renal in terms of neurological function is moderate at best, failure from hyperosmolality, haemolysis and congestive whereas the adverse effects of corticosteroids in these heart failure. 43 patients are clearly substantial. 50 Intravenous mannitol (20–25% solution) has become the mainstay of osmotic therapy in many centres, despite Mannitol is widely used in patients with intracerebral a long-standing lack of evidence from controlled stud- haemorrhage and a depressed level of consciousness, ies. 44 Recently a clinical trial in India randomized 128 to decrease intracranial pressure and alleviate the patients with any type of supratentorial haemorrhage space-occupying effect of the haemorrhage in a within 6 days and found no difference in case fatality deteriorating patient, although this custom is not at 1 month. 45 The correct dose cannot be predicted, but backed up by randomized trials with clinical status a safe regimen is to give 0.75–1 g/kg initially, then rather than pressure as the outcome variable. With 0.25–0.5 g/kg every 3–5 h, depending on clinical re- corticosteroids the benefits have so far not been sponse and osmolality (the aim being 295–310 mosm/L). 46 proved to outweigh the disadvantages. A controlled trial of a high dose (1.4 mg/kg) against a standard dose (0.7 mg/kg) of mannitol in 141 13.3.3 Prevention of deep venous thrombosis preoperative patients with traumatic temporal lobe and pulmonary embolism haemorrhages and ipsilateral pupillary dilatation found better outcome after 6 months in the high-dose group (See also section 11.13.) Patients with ICH have a (61% vs 33%), 47 but the relevance of this trial is substantial risk of deep venous thrombosis (DVT): .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 707 13.3 Management 707 Study Treatment Control Relative risk (fixed) Weight Relative risk (fixed) n/N n/N 95% CI (%) 95% CI At one month Desai 1998 10/12 12/14 17.3 0.97 [0.70, 1.35] Ogun 2001 14/15 11/12 19.1 1.02 [0.82, 1.27] Poungvarin 1987 37/46 41/47 63.5 0.92 [0.77, 1.10] Subtotal (95% CI) 73 73 100.0 0.95 [0.83, 1.09] Total events: 61 (Treatment), 64 (Control) Test for heterogeneity chi-square = 0.52 df = 2 P = 0.77 I = 0.0% 2 Test for overall effect Z = 0.76 P = 0.4 At six months Subtotal (95% CI) 00 0.0 Not estimable Total events: 0 (Treatment), 0 (Control) Test for heterogeneity: not applicable Test for overall effect: not applicable Total (95% CI) 73 73 100.0 0.95 [0.83, 1.09] Total events: 61 (Treatment), 64 (Control) 2 Test for heterogeneity chi-square = 0.52 df = 2 P = 0.77 I = 0.0% Test for overall effect Z = 0.76 P = 0.4 0.1 0.2 0.5 1 2510 Favours treatment Favours control Fig. 13.2 Glucocorticoid treatment vs controls in intracerebral haemorrhage. The same conventions as in Fig. 13.1. The outcome is ‘poor outcome at the end of follow up’. 50 between 30% and 70%, depending on the severity of the pulmonary embolism was significantly lower in the group stroke and the degree of immobilization, and on how with treatment started on day 2, without a concomitant hard one looks for it. The risk of pulmonary embolism increase in rebleeding. But, as the group with early treat- has not been studied specifically for patients with hae- ment was studied after the two other dosage groups, the morrhagic stroke, but extrapolation from what is known results may well have been influenced by factors other for ischaemic stroke or stroke in general the estimate is than the interval until the initiation of treatment. 1–2% in the first month (section 11.13). 51,52 However, physicians are intuitively cautious in using antithrom- Aspirin botic drugs in patients with intracerebral haemorrhages. On the other hand, drug regimens that prevent clotting Aspirin is effective in reducing the risk of DVT (by as a consequence of venous stasis do not necessarily approximately 39%) as well as pulmonary embolism increase the risk of a small artery in the brain ruptur- in an overview of all trials in the perioperative period ing for a second time. The definitive answer should and in medical patients at increased risk (section 11.13). of course come from clinical trials, not theoretical In a systematic review of all published trials comparing considerations. any antithrombotic agent with control among patients with any form of intracranial haemorrhage, there were two randomized trials in which aspirin had been inad- Subcutaneous heparin vertently given (for variable periods) to 398 patients Subcutaneous heparin, usually given as 5000 U two or with acute ICH (and placebo to 375 patients). 56 The odds three times a day, reduces the frequency of DVT by 68%, ratio (OR) for these ICH patients treated with aspirin at least after general, orthopaedic and urological sur- was 0.96 (95% CI, 0.62–1.5) for death, and 1.02 (95% CI, gery. 53 In a small, but so far unique, clinical trial in 46 0.5–1.8) for recurrent haemorrhage. These scant data patients with ICH this regimen was applied 4 days after do not support reliable conclusions about the safety or the onset in the active treatment group and 10 days after otherwise of antithrombotic agents in patients with the haemorrhage in the control group. 54 In the control acute intracerebral haemorrhage. group 40% (9 of 23) developed some evidence of pul- monary embolism (as detected by perfusion scintigraphy Compression stockings and intermittent pneumatic of the lungs) against 22% (5 of 22) in the treated group. compression In a subsequent non-randomized study, in 68 patients with ICH, treatment was started on day 2, 4 or 10 after Physical methods are of course likely to be the safest the haemorrhage. 55 For what it is worth, the risk of method of thrombosis prevention in patients with ICH, .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 708 708 Chapter 13 Specific treatment of intracerebral haemorrhage although occasionally they may cause pressure sores 13.3.4 Treatment of iatrogenic intracerebral (section 11.13). Essentially there are two methods: inter- haemorrhage mittent pneumatic compression, and graduated com- pression stockings which vary the pressure from being Anticoagulants highest at the ankle and decreasing proximally. A sys- tematic review of 16 trials (totalling more than 2200 Anticoagulants are associated with an increased risk patients) of graduated compression stockings in oper- of intracerebral haemorrhage (section 8.4.1). It seems ated or otherwise immobilized patients for a variety of rational to reverse the deficiency of clotting factors as conditions found an odds ratio for deep venous throm- quickly as possible and to accept the risk of a thrombotic bosis of 0.34 (95% CI, 0.25–0.46) if graduated pressure event, especially in view of the frequent progression of stockings was the only treatment, and of 0.24 (95% CI, clinical deficits that occurs in these patients. Under- 0.15–0.37) if they were used on a background of other standably the speed of treatment has never been put to prophylactic measures. 57 In stroke patients alone, two the test in a clinical trial, but observational data confirm small trials (123 patients in all) could not demonstrate that early normalization of clotting status is associated any such effect. 58 with a relatively low rate of haemorrhage enlargement. 61 A later trial tested intermittent pneumatic compres- In current practice the first step is i.v. 10–20 mg vitamin sion, specifically in 151 patients with ICH; all were K, at not more than 5 mg/min, rapidly followed by also fitted with graduated elastic stockings. 59 On day infusion of prothrombin complex concentrate (PCC), 10, DVT (as assessed by compression ultrasonography) which contains the coagulation factors II, VII, IX and was found in 5% of the patients in the group with X; 62 every half hour counts in halting progression of the intermittent compression, against 16% in those with haemorrhage. 63 Infusion of these factors alone restores stockings alone, corresponding to a relative risk of the coagulation system more rapidly than whole plasma 0.29 (95% CI, 0.08–1.00), a marginally significant and is safer from the point of view of transmission of difference. virus particles (section 12.4.7). 64–66 Pulmonary embolism was not included in these sys- When to resume anticoagulants in patients with a tematic reviews, but there is no reason to assume that strong indication for this treatment (e.g. those with the relationship with DVT is different in patients with mechanical heart valves) is a problem many clinicians ICH. Also a significantly protective effect of compression will recognize. Only anecdotal experience is available, stockings on the risk of pulmonary emboli has been such as a series from the Mayo Clinic in which only one demonstrated in a large trial of patients who underwent ischaemic stroke occurred in 52 patients with mechanical cardiothoracic operations. 60 Thus, unless the disadvant- heart valves in whom anticoagulants were discontinued ages of heparin are proved to be merely theoretical, for a median period of 10 days, 67 in contrast to seven application of graduated compression stockings, prefer- similar patients from Heidelberg in three of whom large ably supplemented with intermittent pneumatic com- brain infarcts occurred within 4 days of normalization pression, seems to offer the greatest protection and the of the international normalized ratio (INR). 68 The essen- smallest risk, although the method is fairly labour inten- tial issue in deciding when to restart is probably not sive and the stockings can be uncomfortable after a time. when blood has disappeared from the CT scan (that may To resolve the existing uncertainties, a multicentre trial take weeks or months), but when the ruptured vessel of graduated compression stockings has been initiated can be assumed to have sufficiently healed. Between in the UK (section 11.13). 1–2 weeks after the haemorrhage – depending on the stringency of the indication – is probably not a bad Subcutaneous heparin and the application of 69 guess. No rebleeds were reported with such a regimen, graduated compression stockings both decrease 70,71 although the numbers were small. the risk of deep venous thrombosis in bedridden patients by approximately 70%, and aspirin by approximately 39%, as judged from clinical trials Thrombolytic therapy in a variety of postoperative conditions. Because Thrombolytic therapy after myocardial infarction is the safety of heparin and aspirin is largely unknown complicated by intracerebral haemorrhage in only a in patients with intracerebral haemorrhage, small proportion (section 8.4.3), but the case fatality compression stockings are presently the preferred is high. This grim prognosis warrants attempts at method of prophylaxis, despite being labour intervention, even without the benefits of controlled intensive, and despite the lack of direct evidence clinical trials. These measures include control of any in patients with primary intracerebral hypertension and infusion of coagulation factors; the haemorrhage. use of antifibrinolytic drugs is controversial even from .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 709 13.3 Management 709 a theoretical point of view. Surgical treatment is subsequently abandoned because only small amounts of hazardous, given that amyloid angiopathy is often a clot could be obtained, and because the procedure could contributing factor (section 13.3.7). Life-threatening precipitate ‘blind’ rebleeding. 78,79 arrhythmias may occur as a complication of the myo- cardial infarct for which the treatment was given, often Craniotomy before the onset of the neurological symptoms, and these should of course be promptly treated. 72 Open surgery was the method for removing haematomas In patients with ischaemic stroke who develop major in five trials, one of them (with 131 patients) dating intracerebral haemorrhage after thrombolytic treatment from the pre-CT era. A Cochrane review (last updated (section 12.5.2) there is no evidence about what to do. in 1999) of this early study and two subsequent trials In view of the little we know about operative treatment (totalling 64 randomized patients) in the 1980s showed for intracerebral haemorrhage in general (section 13.3.5) a non-significant trend towards increased risk of death it seems best to leave well alone. The utility of prothrom- and dependency among survivors (odds ratio 1.99, 99% bin complex concentrate or recombinant factor VIIa has CI, 0.92–4.31). Two subsequent but small single-centre not been studied so far. randomized trials (54 patients altogether) showed results that were entirely within the range of the overview. 80,81 A pseudo-randomized trial in Houston tested the optimal Aspirin time for clot evacuation – assuming that the operation Aspirin treatment is associated with a small risk of was indicated at all; in the experimental group patients intracerebral haemorrhage (section 8.4.2). It is reason- were operated within 4 h of symptom onset, patients able to stop the drug once the diagnosis has been operated within 12 h forming the control group. The made, although no disasters have been documented in study was stopped early because of rebleeding in the instances in which aspirin treatment was continued, or ‘ultra-early’ group. 82 even restarted. At any rate the antiplatelet effect lasts for The year 2005 saw publication of a landmark trial several days after discontinuation of the drug. 73 In the (STICH) that has considerably added to the modest trials of aspirin for secondary prevention after cerebral information from controlled studies so far. 83 A total of ischaemia some patients with small intracerebral hae- 1003 patients were randomized, four times as many as in morrhages must have been inadvertently included all previous trials taken together; all had supratentorial before CT scanning established the true diagnosis; there haemorrhage. In the group allocated to early opera- were no obviously untoward effects, but the confidence tion (craniotomy in 75%) the procedure was scheduled interval is still wide (section 12.3.3). 74 In patients on within 24 h, whereas in the other group treatment was regular treatment with aspirin the outcome after ICH is initially conservative. A second notable feature was the relatively poor, 75 but this is explained by confounding introduction of a ‘sliding dichotomy’ for assessing out- factors; in a large observational study from Germany, come, to take account of the methodological problem the effect disappeared after adjustment for age and pre- that expectations about the result of any treatment are morbid condition. 76 higher when the initial condition of the patient is bet- ter. 84 The result was that the cut-off point for ‘favourable outcome’ depended on three prognostic features at base- 13.3.5 Surgical treatment of supratentorial line: level of consciousness on the Glasgow Coma Scale, haemorrhage age and volume of the haemorrhage. For 516 patients The frequency of surgical intervention for intracerebral with a good prognostic profile on admission a favourable haemorrhage varies widely between centres, and even outcome included only the categories ‘good recovery’ between participants in a clinical trial. 77 There are and ‘moderate disability’ on the extended Glasgow out- four possible surgical procedures to treat ICH: simple come scale, whereas for the other half of the study aspiration, craniotomy with open surgery, endoscopic patients with a poorer prognosis ‘severe disability’ was evacuation, and stereotactic aspiration. Unless speci- included in the definition of a favourable outcome. 85 fically indicated otherwise, the assumption is that we are The analysis of outcome after 6 months failed to show mostly dealing with deep haemorrhages, from degener- any benefit of craniotomy: of patients randomized to ative small vessel disease. early surgery, 122 (26%) had a favourable outcome com- pared with 118 (24%) of those randomized to initially conservative treatment; odds ratio 0.89 (95% CI, 0.66– Simple aspiration 1.19). Separate analysis of 12 prespecified subgroups Aspiration, not accompanied by any other interven- showed that operation was perhaps beneficial in patients tion, was attempted mainly in the 1950s but this was with superficial haemorrhages (1 cm or less from the .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 710 710 Chapter 13 Specific treatment of intracerebral haemorrhage cortical surface), but of course this may be chance effect. agents, has been reported in several observational stud- The efficacy of operative treatment did not significantly ies involving hundreds of patients with supratentorial differ between patients with deep haemorrhages (approx- haemorrhage; some of these concentrated on haemor- imately 60%) and those with lobar haemorrhage. rhages with intraventricular extension. 34 Subsequently No systematic review has yet been completed that two controlled trials of this technique have been per- included this major trial with the five small trials per- formed. A Japanese trial randomized 242 patients with formed previously, but of course such an analysis will putaminal haemorrhage and a moderately decreased level not change the current conclusion that – as a routine – of consciousness (eyes closed but opened to stimuli) craniotomy is not the treatment of choice in patients between stereotactic haemorrhage evacuation and con- with intracerebral haemorrhage. A question not addressed servative treatment. Outcome, assessed by a blinded by the STICH trial is whether operation improves out- observer in terms of independence, was better in the come in patients deteriorating because of haemorrhage surgical group. 91 A smaller trial in the Netherlands (70 expansion. A consecutive series of 26 such patients patients) combined stereotactic aspiration with lique- from the Mayo Clinic (of whom 24 had lobar haemor- faction by means of a plasminogen activating substance; rhage) reported that no patient regained independence – no conclusive differences emerged. 92 with or without operation – if corneal and oculocephalic reflexes had been lost, but that approximately one- 13.3.6 Surgical treatment of infratentorial quarter of the remaining group (6 of 21) did regain haemorrhage independence. 86 Cerebellar haemorrhage Given the present state of knowledge routine craniotomy and evacuation of the haemorrhage Cerebellar haemorrhages (Figs 8.29 and 13.3) have fairly is not indicated for patients with supratentorial characteristic clinical features (section 3.3.7), with the haemorrhage, deep or lobar. No subgroup of patients exception of massive haemorrhages, which are clinically has yet been reliably identified for whom operation might be beneficial. Endoscopic evacuation There are enthusiasts for almost any conceivable type of operation. Endoscopic evacuation of brain haemor- rhage by stereotactic methods is no exception. 87,88 The only controlled study so far involved two groups of 50 patients. 89 The authors’ interpretation was that surgery had no effect in putaminal and thalamic haemorrhage, but that it was beneficial in subcortical haemorrhages, provided patients are aged 60 years or less and are alert or somnolent. The problem with this analysis is not only that it depends on subgroups, but also no overall analysis was reported and some outcome categories were not reported at all. Recalculation of the results for all patients according to the proportion who were dead or depend- ent showed an odds ratio of 0.45 in favour of endoscopic operation, but this was not statistically significant (95% CI, 0.19–1.04). 90 In conclusion, stereotactic endoscopic evacuation is a promising technique, but its benefits and specific indications remain to be confirmed by well- conducted and large clinical trials. Stereotactic aspiration with thrombolysis Since the 1990s, stereotactic aspiration without endos- Fig. 13.3 Unenhanced CT scan showing large cerebellar copy, mostly combined with instillation of fibrinolytic haematoma. .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 711 13.3 Management 711 indistinguishable from brainstem strokes, and very shunting to improve the level of consciousness. 105 Never- small haemorrhages which may simulate a peripheral theless we do not wish to deny that ventriculostomy disorder of the vestibular system. 93 For decades there can be the only effective treatment in some patients has been a strong impression that surgical evacuation with appropriate clinical features of hydrocephalus, i.e. saves lives in patients with cerebellar haemorrhages in whom deterioration of consciousness is gradual, with who have clinical evidence of progressive brainstem sustained downward gaze and small unreactive pupils compression. 94 The need for timely surgical treatment in but no signs of pontine compression. 100 patients who are in danger of progressive brainstem compression is dictated on the one hand by the poten- Surgical evacuation of cerebellar haemorrhages can be tially fatal outcome and on the other hand by the often life-saving, often with surprisingly few neurological surprisingly mild sequelae after operation. 95 So strong sequelae. Sound indications for evacuation are the is this notion, that a clinical trial in this category of combination of a depressed level of consciousness patients is as unlikely to be mounted as it would be in with signs of progressive brainstem compression acute appendicitis. This is not to say there are no areas of (unless all brainstem reflexes have been lost for more controversy. than a few hours, in which case a fatal outcome is First, there is no doubt that some patients can be inevitable), or haemorrhage diameter is greater than managed conservatively, but there is uncertainty about 3–4 cm. If the patient has a depressed level of the selection criteria. 96,97 In general, an impaired level consciousness and hydrocephalus, without signs of of consciousness (that is, a Glasgow Coma Scale Score brainstem compression and with a haemorrhage less of 13/15 or less) seems a good indication for operative than 3 cm, ventriculostomy can be carried out as an intervention, provided corneal and oculocephalic reflexes initial (and perhaps only) procedure. have not been lost, in which case the outcome is invari- ably fatal. 98 Some neurosurgeons advocate surgery with As with supratentorial haemorrhage, some patients large haemorrhages (>3–4 cm in diameter), even in alert with cerebellar haemorrhage have been successfully patients, on the basis of experience that delayed deter- treated by stereotactic aspiration, with or without instil- ioration of consciousness may be so rapid that the patient lation of fibrinolytic drugs. 106 Any advantage of these cannot be salvaged at this later stage. 99 Others maintain techniques over the conventional approach with sub- that such rapid deterioration depends not so much on occipital craniotomy remains to be proved. The same the absolute size of the haemorrhage as on the degree applies to ventriculostomy by endoscopic perforation of of effacement of the fourth ventricle and that operat- the third ventricle towards the basal cisterns, 107 rather ive evacuation is indicated in all patients in whom than by the conventional method of a catheter inserted the fourth ventricle is completely obliterated. 100 Since in the lateral ventricle. effacement of the fourth ventricle is associated with hydrocephalus and location of the haemorrhage in the Pontine haemorrhage midline (vermis), it is not surprising that these features also predispose to secondary deterioration. 101 A rare Pontine haemorrhages (Figs 8.31 and 13.4) are not as category of cerebellar haemorrhages are those associated invariably fatal as when the diagnosis could only be with neurosurgical operations (for any lesion) in the made at postmortem, but the case fatality is still around supratentorial compartment, so-called ‘remote cerebellar 50%. 17,108 The outcome depends to an important extent haemorrhages’. 102 The pathogenesis is unclear, while the on the cause, since haemorrhage in the pons from a clinical course is almost invariably self-limiting. cavernous malformation is rarely fatal. 109 Cavernous A second contentious issue is that in the past some malformations are separately dealt with below (section neurosurgeons have argued that it is not direct compres- 13.3.8). The management of patients with ‘hypertensive’ sion of the brainstem but obstructive hydrocephalus pontine haemorrhage is usually conservative, but some that is the main problem, and that ventriculostomy is a case reports have documented successful stereotactic sufficient measure to prevent a fatal outcome. 103 How- aspiration. 110 The natural history may or may not have ever, there are two major objections against ventricular been influenced by these interventions, and surgical shunting as a panacea for cerebellar haemorrhages. The failures in similar cases are likely to receive rather less first is the rapid disappearance of pupillary, corneal publicity. There is an extremely narrow margin of uncer- and oculocephalic reflexes testifying brainstem compres- tainty between patients with poor prognosis because of sion in some patients for whom operation is too late, absent corneal and oculocephalic reflexes and those with something not seen with acute hydrocephalus from other small haemorrhages who will do well with conservative causes. 104 The second is the frequent failure of ventricular management. .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 712 712 Chapter 13 Specific treatment of intracerebral haemorrhage There is little doubt that (recurrent) haemorrhages from cavernous malformations are less destructive than brainstem haemorrhages from an arterial source. 109 Also the lesions may spontaneously regress. 117 Successes of surgical treatment make it to a publication, 118 but the failure rate can only be guessed at. At any rate incom- plete removal can prompt rebleeding. 119 Stereotactic cobalt-generated radiation (‘gamma knife’) has been applied to a large number of patients with cav- ernous malformations; there is no controlled evidence for the efficacy of this treatment and nothing more than a suggestion that seizures or episodes of rebleeding are less likely to occur afterwards. 120,121 Cavernous malformations, whether in a cerebral hemisphere or in the brainstem, carry a moderate risk of (re)bleeding, while the functional deficits from such haemorrhages are limited. Surgical treatment or stereotactic radiation is regularly performed but there Fig. 13.4 Unenhanced CT brain scan showing a pontine are no randomized controlled studies. haematoma (arrows) in a patient who walked, unsteadily, into (and out of) the outpatient clinic. 13.3.9 Treatment of dural arteriovenous fistulae These malformations are heterogeneous, in that they 13.3.7 Treatment of lobar haemorrhage from may or may not be secondary to occlusion of a major (presumed) amyloid angiopathy sinus and that the abnormal venous drainage from In patients with lobar haemorrhage from presumed amy- meningeal arteries may be channelled into a dural sinus loid angiopathy and an impaired level of consciousness or into superficial veins on the surface of the brain, the prognosis is often poor after surgical intervention. 111 cerebellum, or brainstem (section 8.2.8). Accordingly, Uncontrolled series with a large proportion of survivors there is a large variety of methods used to occlude the after operative treatment fail to answer the question fistula. Surgical techniques include selective ligation of whether the outcome would have been worse without leptomeningeal draining veins, 122,123 resection of fistu- intervention. 112,113 Given these uncertainties and the lous sinus tracts, 122 and a cranial base approach with danger that the operation provokes new haemorrhages extradural bone removal. 124 Endovascular techniques from brittle vessels at distant sites, 114 conservative treat- may consist of an approach from the arterial or venous ment seems the best option unless a deteriorating level side, 125,126 recanalization and stenting of a venous of consciousness justifies the risk of operation. sinus, 127 or venous embolization via a craniotomy. 128 Finally stereotactic cobalt-generated radiation (‘gamma knife’) is also used for obliterating arteriovenous 13.3.8 Treatment of cavernous malformations 129,130 fistulae, sometimes in conjunction with an For cavernous malformations in a cerebral hemisphere endovascular approach. 131 or in the cerebellum, any intervention is exceptional, because recurrent haemorrhages from these lesions are 13.3.10 Treatment of moyamoya syndrome usually limited in severity (section 8.2.5). Series with associated with haemorrhage operated patients have been published, 115 but these underline rather than answer the question whether sur- Gradual stenosis or occlusion of the terminal portions gical treatment really improves on the natural history. of the internal carotid arteries or proximal middle cere- The management of cavernous malformations in the bral arteries, most often the result of an idiopathic pro- brainstem is also controversial, since the natural history liferative process of the endothelial layer, may lead to the often is rather favourable, regardless of location. 116 Prob- formation of an abnormal collateral network of fragile ably pressure from patients themselves accounts for collaterals, which occasionally rupture (sections 8.2.12 some of the operations. and 7.5). To relieve the pressure on the collaterals by .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 713 References 713 constructing a bypass such as between the superficial 13 Hays A, Diringer MN. Elevated troponin levels are temporal artery and the middle cerebral artery is a logical associated with higher mortality following intracerebral idea, 132 but whether this indeed prevents (re)bleeding hemorrhage. Neurology 2006; 66:1330–4. remains uncertain. In fact an aneurysm may form and 14 Roquer J, Rodriguez CA, Gomis M, Ois A, Munteis E, rupture at the site of the arterial bypass. 133 Bohm P. Serum lipid levels and in-hospital mortality in patients with intracerebral hemorrhage. Neurology 2005; 65:1198–202. 15 Flemming KD, Wijdicks EFM, Li HZ. Can we predict poor outcome at presentation in patients with lobar hemorrhage? Cerebrovasc Dis 2001; 11:183–9. References 16 Liliang PC, Liang CL, Lu CH et al. Hypertensive caudate hemorrhage prognostic predictor, outcome, and role of external ventricular drainage. Stroke 2001; 32:1195–200. 17 Wijdicks EFM, St. Louis E. Clinical profiles predictive of 1 Rosand J, Eskey C, Chang Y, Gonzalez RG, Greenberg outcome in pontine hemorrhage. Neurology 1997; SM, Koroshetz WJ. Dynamic single-section CT 49:1342–6. demonstrates reduced cerebral blood flow in acute 18 Davis SM, Broderick J, Hennerici M et al. Hematoma intracerebral hemorrhage. Cerebrovasc Dis 2002; growth is a determinant of mortality and poor outcome 14:214–20. after intracerebral hemorrhage. Neurology 2006; 2 Castillo J, Dávalos A, Alvarez-Sabín J et al. Molecular 66:1175–81. signatures of brain injury after intracerebral hemorrhage. 19 Schwarz S, Häfner K, Aschoff A, Schwab S. Incidence and Neurology 2002; 58:624–9. prognostic significance of fever following intracerebral 3 Butcher KS, Baird T, MacGregor L, Desmond P, Tress B, hemorrhage. Neurology 2000; 54:354–61. Davis S. Perihematomal edema in primary intracerebral 20 Gebel JM, Jr., Jauch EC, Brott TG et al. Relative edema hemorrhage is plasma derived. Stroke 2004; volume is a predictor of outcome in patients with 35:1879–85. hyperacute spontaneous intracerebral hemorrhage. 4 Gebel JM, Jr., Jauch EC, Brott TG et al. Natural history of Stroke 2002; 33:2636–41. perihematomal edema in patients with hyperacute 21 Ronning OM, Guldvog B, Stavem K. The benefit of spontaneous intracerebral hemorrhage. Stroke 2002; an acute stroke unit in patients with intracranial 33:2631–5. haemorrhage: a controlled trial. J Neurol Neurosurg 5 Siddique MS, Fernandes HM, Wooldridge TD, Fenwick JD, Psychiatry 2001; 70:631–4. Slomka P, Mendelow AD. Reversible ischemia around 22 Friederich PW, Henny CP, Messelink EJ et al. Effect of intracerebral hemorrhage: a single-photon emission recombinant activated factor VII on perioperative blood computerized tomography study. J Neurosurg 2002; loss in patients undergoing retropubic prostatectomy: a 96:736–41. double-blind placebo-controlled randomised trial. Lancet 6 Bhattathiri PS, Gregson B, Prasad KS, Mendelow AD. 2003; 361:201–5. Intraventricular hemorrhage and hydrocephalus after 23 You H, Al-Shahi R. Haemostatic drug therapies for acute spontaneous intracerebral hemorrhage: results from the primary intracerebral haemorrhage. Cochrane Database STICH trial. Acta Neurochir Suppl 2006; 96:65–8. Syst Rev 2006; CD005951. 7 Dennis MS. Outcome after brain haemorrhage. 24 Earnshaw SR, Joshi AV, Wilson MR, Rosand J. Cost- Cerebrovasc Dis 2003; 16(Suppl 1):9–13. effectiveness of recombinant activated factor VII in the 8 Flaherty ML, Haverbusch M, Sekar P et al. Long-term treatment of intracerebral hemorrhage. Stroke 2006; mortality after intracerebral hemorrhage. Neurology 2006; 37:2751–8. 66:1182–6. 25 Tomson J, Lip GY. Blood pressure changes in acute 9 Ariesen MJ, Algra A, Van der Worp HB, Rinkel GJE. haemorrhagic stroke. Blood Press Monit 2005; 10:197–9. Applicability and relevance of models that predict short 26 Ohwaki K, Yano E, Nagashima H, Hirata M, Nakagomi T, term outcome after intracerebral haemorrhage. J Neurol Tamura A. Blood pressure management in acute Neurosurg Psychiatry 2005; 76:839–44. intracerebral hemorrhage: relationship between elevated 10 Weimar C, Benemann J, Diener HC, for the German Stroke blood pressure and hematoma enlargement. Stroke 2004; Study Collaboration. Development and validation of the 35:1364–7. Essen Intracerebral Haemorrhage Score. J Neurol Neurosurg 27 Qureshi AI, Mohammad YM, Yahia AM et al. A prospective Psychiatry 2006; 77:601–5. multicenter study to evaluate the feasibility and safety of 11 Becker KJ, Baxter AB, Cohen WA et al. Withdrawal of aggressive antihypertensive treatment in patients with support in intracerebral hemorrhage may lead to self- acute intracerebral hemorrhage. J Intensive Care Med 2005; fulfilling prophecies. Neurology 2001; 56:766–72. 20:34–42. 12 Delgado P, Alvarez-Sabin J, Abilleira S et al. Plasma d-dimer 28 Arakawa S, Saku Y, Ibayashi S, Nagao T, Fujishima M. predicts poor outcome after acute intracerebral Blood pressure control and recurrence of hypertensive hemorrhage. Neurology 2006; 67:94–8. brain hemorrhage. Stroke 1998; 29:1806–9. .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 714 714 Chapter 13 Specific treatment of intracerebral haemorrhage 29 Broderick JP, Adams HP, Jr., Barsan W et al. Guidelines 46 Dziedzic T, Szczudlik A, Klimkowicz A, Rog TM, Slowik A. for the management of spontaneous intracerebral Is mannitol safe for patients with intracerebral hemorrhage: a statement for healthcare professionals from hemorrhages? Renal considerations. Clin Neurol Neurosurg a special writing group of the Stroke Council, American 2003; 105:87–9. Heart Association. Stroke 1999; 30:905–15. 47 Cruz J, Minoja G, Okuchi K. Major clinical and 30 Guyot LL, Dowling C, Diaz FG, Michael DB. Cerebral physiological benefits of early high doses of mannitol monitoring devices: analysis of complications. Acta for intraparenchymal temporal lobe hemorrhages Neurochir Suppl 1998; 71:47–9. with abnormal pupillary widening: a randomized trial. 31 Cremer OL, van Dijk GW, van Wensen E et al. Effect of Neurosurgery 2002; 51:628–37. intracranial pressure monitoring and targeted intensive 48 Righetti E, Celani MG, Cantisani T, Sterzi R, Boysen G, care on functional outcome after severe head injury. Crit Ricci S. Glycerol for acute stroke. Cochrane Database Syst Care Med 2005; 33:2207–13. Rev 2004; CD000096. 32 Adams RE, Diringer MN. Response to external ventricular 49 Ogden AT, Mayer SA, Connolly ES, Jr. Hyperosmolar drainage in spontaneous intracerebral hemorrhage with agents in neurosurgical practice: the evolving role of hydrocephalus. Neurology 1998; 50:519–23. hypertonic saline. Neurosurgery 2005; 57:207–15. 33 Holtzman RN, Brust JC, Ainyette IG et al. Acute ventricular 50 Feigin VL, Anderson N, Rinkel GJ, Algra A, van Gijn J, hemorrhage in adults with hydrocephalus managed by Bennett DA. Corticosteroids for aneurysmal subarachnoid corpus callosotomy and fenestration of the septum haemorrhage and primary intracerebral haemorrhage. pellucidum. Report of three cases. J Neurosurg 2001; Cochrane Database Syst Rev 2005; CD004583. 95:111–15. 51 Davenport RJ, Dennis MS, Wellwood I, Warlow CP. 34 Nieuwkamp DJ, de Gans K, Rinkel GJE, Algra A. Treatment Complications after acute stroke. Stroke 1996; 27:415–20. and outcome of severe intraventricular extension in 52 Kelly J, Hunt BJ, Lewis RR, Rudd A. Anticoagulation or patients with subarachnoid or intracerebral hemorrhage: inferior vena cava filter placement for patients with a systematic review of the literature. J Neurol 2000; primary intracerebral hemorrhage developing venous 247:117–21. thromboembolism? Stroke 2003; 34:2999–3005. 35 Lapointe M, Haines S. Fibrinolytic therapy for 53 Collins R, Scrimgeour A, Yusuf S, Peto R. Reduction in intraventricular hemorrhage in adults. Cochrane Database fatal pulmonary embolism and venous thrombosis by Syst Rev 2002; CD003692. perioperative administration of subcutaneous heparin. 36 Lee MW, Pang KY, Ho WW, Wong CK. Outcome analysis of Overview of results of randomized trials in general, intraventricular thrombolytic therapy for intraventricular orthopedic, and urologic surgery. N Engl J Med 1988; haemorrhage. Hong Kong Med J 2003; 9:335–40. 318:1162–73. 37 Fountas KN, Kapsalaki EZ, Parish DC et al. Intraventricular 54 Dickmann U, Voth E, Schicha H, Henze T, Prange H, administration of rt-PA in patients with intraventricular Emrich D. Heparin therapy, deep-vein thrombosis and hemorrhage. South Med J 2005; 98:767–73. pulmonary embolism after intracerebral hemorrhage. 38 Stocchetti N, Maas AI, Chieregato A, van der Plas AA. Klin Wochenschr 1988; 66:1182–3. Hyperventilation in head injury: a review. Chest 2005; 55 Boeer A, Voth E, Henze T, Prange HW. Early heparin 127:1812–27. therapy in patients with spontaneous intracerebral 39 Muizelaar JP, Marmarou A, Ward JD et al. Adverse effects haemorrhage. J Neurol Neurosurg Psychiatry 1991; of prolonged hyperventilation in patients with severe 54:466–7. head injury: a randomized clinical trial. J Neurosurg 1991; 56 Keir SL, Lewis SC, Wardlaw JM, Sandercock PA, on behalf 75:731–9. of IST and CAST collaborative groups. Effect of aspirin or 40 Schierhout G, Roberts I. Hyperventilation therapy for heparin inadvertently given to patients with hemorrhagic acute traumatic brain injury. Cochrane Database Syst Rev stroke. Stroke 2000; 31:314. 2000; CD000566. 57 Amaragiri SV, Lees TA. Elastic compression stockings for 41 Nau R. Osmotherapy for elevated intracranial pressure: prevention of deep vein thrombosis. Cochrane Database a critical reappraisal. Clin Pharmacokinet 2000; Syst Rev 2000; CD001484. 38:23–40. 58 Mazzone C, Chiodo GF, Sandercock P, Miccio M, Salvi R. 42 Bettmann MA. Frequently asked questions: iodinated Physical methods for preventing deep vein thrombosis in contrast agents. Radiographics 2004; 24(Suppl 1):S3–10. stroke. Cochrane Database Syst Rev 2004; CD001922. 43 Yu YL, Kumana CR, Lauder IJ et al. Treatment of acute 59 Lacut K, Bressollette L, Le Gal G et al. Prevention of venous cerebral hemorrhage with intravenous glycerol. A double- thrombosis in patients with acute intracerebral blind, placebo-controlled, randomized trial. Stroke 1992; hemorrhage. Neurology 2005; 65:865–9. 23:967–71. 60 Ramos R, Salem BI, De Pawlikowski MP, Coordes C, 44 Bereczki D, Liu M, do Prado GF, Fekete I. Mannitol for Eisenberg S, Leidenfrost R. The efficacy of pneumatic acute stroke. Cochrane Database Syst Rev 2001; CD001153. compression stockings in the prevention of pulmonary 45 Misra UK, Kalita J, Ranjan P, Mandal SK. Mannitol in embolism after cardiac surgery. Chest 1996; 109:82–5. intracerebral hemorrhage: a randomized controlled study. 61 Yasaka M, Minematsu K, Naritomi H, Sakata T, J Neurol Sci 2005; 234:41–5. Yamaguchi T. Predisposing factors for enlargement .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 715 References 715 of intracerebral hemorrhage in patients treated with primary intracerebral hemorrhage is an independent warfarin. Thromb Haemost 2003; 89:278–83. predictor for death. Stroke 2006; 37:129–33. 62 Steiner T, Rosand J, Diringer M. Intracerebral hemorrhage 76 Foerch C, Sitzer M, Steinmetz H, Neumann-Haefelin T. associated with oral anticoagulant therapy: current Pretreatment with antiplatelet agents is not independently practices and unresolved questions. Stroke 2006; associated with unfavorable outcome in intracerebral 37:256–62. hemorrhage. Stroke 2006; 37:2165–7. 63 Goldstein JN, Thomas SH, Frontiero V et al. Timing of 77 Gregson BA, Mendelow AD. International variations fresh frozen plasma administration and rapid correction in surgical practice for spontaneous intracerebral of coagulopathy in warfarin-related intracerebral hemorrhage. Stroke 2003; 34:2593–7. hemorrhage. Stroke 2006; 37:151–5. 78 McKissock W, Richardson A, Walsh L. Primary 64 Makris M, Greaves M, Phillips WS, Kitchen S, Rosendaal intracerebral haematoma: results of surgical treatment FR, Preston FE. Emergency oral anticoagulant reversal: the in 244 consecutive cases. Lancet 1959; ii:683–6. relative efficacy of infusions of fresh frozen plasma and 79 Mitsuno T, Kanaya H, Shirakata S, Ohsawa K, Ishikawa Y. clotting factor concentrate on correction of the Surgical treatment of hypertensive intracerebral coagulopathy. Thromb Haemost 1997; 77:477–80. hemorrhage. J Neurosurg 1966; 24:70–6. 65 Butler AC, Tait RC. Management of oral anticoagulant- 80 Morgenstern LB, Frankowski RF, Shedden P, Pasteur W, induced intracranial haemorrhage. Blood Rev 1998; Grotta JC. Surgical treatment for intracerebral hemorrhage 12:35–44. (STICH): a single-center, randomized clinical trial. 66 Huttner HB, Schellinger PD, Hartmann M et al. Hematoma Neurology 1998; 51:1359–63. growth and outcome in treated neurocritical care patients 81 Zuccarello M, Brott T, Derex L et al. Early surgical with intracerebral hemorrhage related to oral treatment for supratentorial intracerebral hemorrhage: anticoagulant therapy: comparison of acute treatment a randomized feasibility study. Stroke 1999; 30:1833–9. strategies using vitamin K, fresh frozen plasma, and 82 Morgenstern LB, Demchuk AM, Kim DH, Frankowski RF, prothrombin complex concentrates. Stroke 2006; Grotta JC. Rebleeding leads to poor outcome in ultra-early 37:1465–70. craniotomy for intracerebral hemorrhage. Neurology 2001; 67 Phan TG, Koh M, Wijdicks EFM. Safety of discontinuation 56:1294–9. of anticoagulation in patients with intracranial 83 Mendelow AD, Gregson BA, Fernandes HM et al. Early hemorrhage at high thromboembolic risk. surgery versus initial conservative treatment in patients Arch Neurol 2000; 57:1710–13. with spontaneous supratentorial intracerebral 68 Bertram M, Bonsanto M, Hacke W, Schwab S. Managing haematomas in the International Surgical Trial in the therapeutic dilemma: patients with spontaneous Intracerebral Haemorrhage (STICH): a randomised trial. intracerebral hemorrhage and urgent need for Lancet 2005; 365:387–97. anticoagulation. J Neurol 2000; 247:209–14. 84 Murray GD, Barer D, Choi S et al. Design and analysis of 69 Estol CJ, Kase CS. Need for continued use of anticoagulants phase III trials with ordered outcome scales: the concept after intracerebral hemorrhage. Curr Treat Options of the sliding dichotomy. J Neurotrauma 2005; 22:511–17. Cardiovasc Med 2003; 5:201–9. 85 Mendelow AD, Teasdale GM, Barer D, Fernandes HM, 70 Wijdicks EFM, Schievink WI, Brown RD, Mullany CJ. The Murray GD, Gregson BA. Outcome assignment in the dilemma of discontinuation of anticoagulation therapy for International Surgical Trial of Intracerebral Haemorrhage. patients with intracranial hemorrhage and mechanical Acta Neurochir (Wien) 2003; 145:679–81. heart valves. Neurosurgery 1998; 42:769–73. 86 Rabinstein AA, Atkinson JL, Wijdicks EFM. Emergency 71 Leker RR, Abramsky O. Early anticoagulation in patients craniotomy in patients worsening due to expanded with prosthetic heart valves and intracerebral hematoma. cerebral hematoma: to what purpose? Neurology 2002; Neurology 1998; 50:1489–91. 58:1367–72. 72 Sloan MA, Price TR, Petito CK et al. Clinical features and 87 Qiu Y, Lin Y, Tian X, Luo Q. Hypertensive intracranial pathogenesis of intracerebral hemorrhage after rt-PA and hematomas: endoscopic-assisted keyhole evacuation and heparin therapy for acute myocardial infarction: The application of patent viewing dissector. Chin Med J (Engl) Thrombolysis in Myocardial Infarction (TIMI) II pilot and 2003; 116:195–9. randomized clinical trial combined experience. Neurology 88 Nishihara T, Nagata K, Tanaka S et al. Newly developed 1995; 45:649–58. endoscopic instruments for the removal of intracerebral 73 Patrono C. Aspirin as an antiplatelet drug. N Engl J Med hematoma. Neurocrit Care 2005; 2:67–74. 1994; 330:1287–94. 89 Auer LM, Deinsberger W, Niederkorn K et al. Endoscopic 74 Keir SL, Wardlaw JM, Sandercock PAG, Chen ZM. surgery versus medical treatment for spontaneous Antithrombotic therapy in patients with any form of intracerebral hematoma: a randomized study. J Neurosurg intracranial haemorrhage: a systematic review of the 1989; 70:530–5. available controlled studies. Cerebrovasc Dis 2002; 90 Hankey GJ, Hon C. Surgery for primary intracerebral 14:197–206. hemorrhage: Is it safe and effective? A systematic review 75 Saloheimo P, Ahonen M, Juvela S, Pyhtinen J, Savolainen of case series and randomized trials. Stroke 1997; ER, Hillbom M. Regular aspirin-use preceding the onset of 28:2126–32. .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 716 716 Chapter 13 Specific treatment of intracerebral haemorrhage 91 Hattori N, Katayama Y, Maya Y, Gatherer A. Impact of 108 Balci K, Asil T, Kerimoglu M, Celik Y, Utku U. Clinical and stereotactic hematoma evacuation on activities of daily neuroradiological predictors of mortality in patients with living during the chronic period following spontaneous primary pontine hemorrhage. Clin Neurol Neurosurg 2005; putaminal hemorrhage: a randomized study. J Neurosurg 108:36–9. 2004; 101:417–20. 109 Rabinstein AA, Tisch SH, McClelland RL, Wijdicks EFM. 92 Teernstra OP, Evers SM, Lodder J, Leffers P, Franke CL, Cause is the main predictor of outcome in patients Blaauw G. Stereotactic treatment of intracerebral with pontine hemorrhage. Cerebrovasc Dis 2004; hematoma by means of a plasminogen activator: 17:66–71. a multicenter randomized controlled trial (SICHPA). 110 Shitamichi M, Nakamura J, Sasaki T, Suematsu K, Stroke 2003; 34:968–74. Tokuda S. Computed tomography guided stereotactic 93 Jensen MB, St Louis EK. Management of acute cerebellar aspiration of pontine hemorrhages. Stereotact Funct stroke. Arch Neurol 2005; 62:537–44. Neurosurg 1990; 54–55:453–6. 94 Fisher CM, Picard EH, Polak A, Dalal P, Ojemann R. Acute 111 McCarron MO, Nicoll JA, Love S, Ironside JW. Surgical hypertensive cerebellar hemorrhage. J Nerv Ment Dis 1965; intervention, biopsy and APOE genotype in cerebral 140:38–57. amyloid angiopathy-related haemorrhage. Br J Neurosurg 95 Cohen ZR, Ram Z, Knoller N, Peles E, Hadani M. 1999; 13:462–7. Management and outcome of non-traumatic cerebellar 112 Greene GM, Godersky JC, Biller J, Hart MN, Adams HP, haemorrhage. Cerebrovasc Dis 2002; 14:207–13. Jr. Surgical experience with cerebral amyloid angiopathy. 96 Wijdicks EFM, Louis EKS, Atkinson JD, Li HZ. Clinician’s Stroke 1990; 21:1545–9. biases toward surgery in cerebellar hematomas: an analysis 113 Izumihara A, Ishihara T, Iwamoto N, Yamashita K, of decision-making in 94 patients. Cerebrovasc Dis 2000; Ito H. Postoperative outcome of 37 patients with lobar 10:93–6. intracerebral hemorrhage related to cerebral amyloid 97 Wessels PH, Ter Berg JWM, Spincemaille GH, Dippel DWJ. angiopathy. Stroke 1999; 30:29–33. Treatment of cerebellar hematoma in the Netherlands: 114 Brisman MH, Bederson JB, Sen CN, Germano IM, a questionnaire survey. Cerebrovasc Dis 2001; 11:190–4. Moore F, Post KD. Intracerebral hemorrhage occurring 98 St Louis EK, Wijdicks EFM, Li HZ, Atkinson JD. Predictors remote from the craniotomy site. Neurosurgery 1996; of poor outcome in patients with a spontaneous cerebellar 39:1114–21. hematoma. Can J Neurol Sci 2000; 27:32–6. 115 Zotta D, Di RA, Scogna A, Ricci A, Ricci G, Galzio RJ. 99 Kobayashi S, Sato A, Kageyama Y, Nakamura H, Supratentorial cavernomas in eloquent brain areas: Watanabe Y, Yamaura A. Treatment of hypertensive application of neuronavigation and functional MRI in cerebellar hemorrhage: surgical or conservative operative planning. J Neurosurg Sci 2005; 49:13–9. management? Neurosurgery 1994; 34:246–50. 116 Kupersmith MJ, Kalish H, Epstein F et al. Natural history 100 Kirollos RW, Tyagi AK, Ross SA, Van Hille PT, Marks PV. of brainstem cavernous malformations. Neurosurgery 2001; Management of spontaneous cerebellar hematomas: 48:47–53. a prospective treatment protocol. Neurosurgery 2001; 117 Yasui T, Komiyama M, Iwai Y et al. A brainstem cavernoma 49:1378–86. demonstrating a dramatic, spontaneous decrease in size 101 St Louis EK, Wijdicks EFM, Li HZ. Predicting neurologic during follow-up: case report and review of the literature. deterioration in patients with cerebellar hematomas. Surg Neurol 2005; 63:170–3. Neurology 1998; 51:1364–9. 118 Ferroli P, Sinisi M, Franzini A, Giombini S, Solero CL, 102 Amini A, Osborn AG, McCall TD, Couldwell WT. Remote Broggi G. Brainstem cavernomas: long-term results of cerebellar hemorrhage. Am J Neuroradiol 2006; 27:387–90. microsurgical resection in 52 patients. Neurosurgery 2005; 103 Shenkin HA, Zavala H. Cerebellar strokes: mortality, 56:1203–12. surgical indications, and results of ventricular drainage. 119 Kikuta K, Nozaki K, Takahashi JA, Miyamoto S, Kikuchi H, Lancet 1982; ii:429–31. Hashimoto N. Postoperative evaluation of microsurgical 104 Van Gijn J, Hijdra A, Wijdicks EFM, Vermeulen M, resection for cavernous malformations of the brainstem. van Crevel H. Acute hydrocephalus after aneurysmal J Neurosurg 2004; 101:607–12. subarachnoid hemorrhage. J Neurosurg 1985; 63:355–62. 120 Régis J, Bartolomei F, Kida Y et al. Radiosurgery for epilepsy 105 Mathew P, Teasdale G, Bannan A, Oluoch-Olunya D. associated with cavernous malformation: retrospective Neurosurgical management of cerebellar haematoma study in 49 patients. Neurosurgery 2000; 47:1091–7. and infarct. J Neurol Neurosurg Psychiatry 1995; 121 Liu KD, Chung WY, Wu HM et al. Gamma knife surgery 59:287–92. for cavernous hemangiomas: an analysis of 125 patients. 106 Mohadjer M, Eggert R, May J, Mayfrank L. CT-guided J Neurosurg 2005; 102(Suppl):81–6. stereotactic fibrinolysis of spontaneous and hypertensive 122 Collice M, D’Aliberti G, Arena O, Solaini C, Fontana RA, cerebellar hemorrhage: long-term results. J Neurosurg 1990; Talamonti C. Surgical treatment of intracranial dural 73:217–22. arteriovenous fistulae: role of venous drainage. 107 Roux FE, Boetto S, Tremoulet M. Third Neurosurgery 2000; 47:56–66. ventriculocisternostomy in cerebellar haematomas. Acta 123 Van Dijk JM, TerBrugge KG, Willinsky RA, Wallace MC. Neurochir (Wien) 2002; 144:337–42. Selective disconnection of cortical venous reflux as .. ..
9781405127660_4_013.qxd 10/13/07 10:54 AM Page 717 References 717 treatment for cranial dural arteriovenous fistulas. 129 O’Leary S, Hodgson TJ, Coley SC, Kemeny AA, Radatz MW. J Neurosurg 2004; 101:31–5. Intracranial dural arteriovenous malformations: results of 124 Kattner DA, Roth TC, Giannotta SL. Cranial base stereotactic radiosurgery in 17 patients. Clin Oncol (R Coll approaches for the surgical treatment of aggressive Radiol) 2002; 14:97–102. posterior fossa dural arteriovenous fistulae with 130 Pan DH, Chung WY, Guo WY et al. Stereotactic leptomeningeal drainage: report of four technical cases. radiosurgery for the treatment of dural arteriovenous Neurosurgery 2002; 50:1156–60. fistulas involving the transverse-sigmoid sinus. J Neurosurg 125 Klisch J, Huppertz HJ, Spetzger U, Hetzel A, Seeger W, 2002; 96:823–9. Schumacher M. Transvenous treatment of carotid 131 Friedman JA, Pollock BE, Nichols DA, Gorman DA, cavernous and dural arteriovenous fistulae: results for Foote RL, Stafford SL. Results of combined stereotactic 31 patients and review of the literature. Neurosurgery radiosurgery and transarterial embolization for dural 2003; 53:836–56. arteriovenous fistulas of the transverse and sigmoid 126 Tomak PR, Cloft HJ, Kaga A, Cawley CM, Dion J, sinuses. J Neurosurg 2001; 94:886–91. Barrow DL. Evolution of the management of tentorial 132 Kawaguchi S, Okuno S, Sakaki T. Effect of direct arterial dural arteriovenous malformations. Neurosurgery 2003; bypass on the prevention of future stroke in patients with 52:750–60. the hemorrhagic variety of moyamoya disease. J Neurosurg 127 Murphy KJ, Gailloud P, Venbrux A, Deramond H, Hanley 2000; 93:397–401. D, Rigamonti D. Endovascular treatment of a grade IV 133 Nishimoto T, Yuki K, Sasaki T, Murakami T, Kodama Y, transverse sinus dural arteriovenous fistula by sinus Kurisu K. A ruptured middle cerebral artery aneurysm recanalization, angioplasty, and stent placement: originating from the site of anastomosis 20 years after technical case report. Neurosurgery 2000; 46:497–500. extracranial-intracranial bypass for moyamoya disease: 128 Houdart E, Saint-Maurice JP, Chapot R et al. Transcranial case report. Surg Neurol 2005; 64:261–5. approach for venous embolization of dural arteriovenous fistulas. J Neurosurg 2002; 97:280–6. .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 719 14 Specific treatment of aneurysmal subarachnoid haemorrhage 14.1 General principles 719 14.2 Treatable causes of poor clinical condition on admission and their management 721 14.3 General care 725 14.4 Prevention of rebleeding 728 14.5 Prevention of delayed cerebral ischaemia 735 14.6 Management of rebleeding 742 14.7 Management of delayed cerebral ischaemia 744 14.8 Management of acute hydrocephalus 746 14.9 Management of systemic complications 749 14.10 Late sequelae and complications 752 discharge patients with a perimesencephalic haemor- 14.1 General principles rhage after 2 or 3 days, and do not give medications other than analgesics. In contrast, patients with an aneurysmal pattern of haemorrhage, even without an aneurysm (or The essence of managing patients with subarachnoid other cause) detected, are admitted for at least 14 days haemorrhage (SAH) from a ruptured aneurysm is decept- and treated similarly to patients who have had an ively simple: make the diagnosis, locate the aneurysm aneurysm detected. and occlude it. But, although occlusion of the ruptured The case fatality after aneurysmal SAH is still as high as 1 aneurysm by an interventional radiologist or neuro- 50%, if one takes into account the 10–15% who die 2 surgeon is an important part, it is by no means the only before receiving medical attention. And of the 50% who part in the management. Treating patients with aneur- survive the initial weeks after the haemorrhage, again ysmal SAH requires taking care of acutely ill patients half are left with permanent deficits in high-level (neuro- threatened by many more complications than only psychological) functioning, resulting in impairment in 3 rebleeding, and should be done by a dedicated, multi- their social role. To a large extent this poor outcome disciplinary team. is caused by the many complications other than rebleed- This chapter deals with the treatment of patients ing that beset the course of the disease: delayed cerebral with a (suspected) aneurysmal cause of their SAH. Those ischaemia, hydrocephalus, a variety of systemic dis- 4 with perimesencephalic non-aneurysmal haemorrhage orders, and complications of the aneurysm treatment 5 are not threatened by complications except for the rare itself. However, despite the considerable proportion of patient who develops a symptomatic hydrocephalus patients who die from the impact of the first haemor- within the first 1 or 2 days after the onset. We usually rhage and the many complications that can occur, the prognosis for patients with SAH has gradually improved over the last decades. 1,6 Patients should always be transferred as quickly as pos- sible to a centre where a multidisciplinary team is avail- Stroke: practical management, 3rd edition. C. Warlow, J. van Gijn, able 24 hours a day, 7 days a week, and where enough M. Dennis, J. Wardlaw, J. Bamford, G. Hankey, P. Sandercock, G. Rinkel, P. Langhorne, C. Sudlow and P. Rothwell. Published patients are managed to maintain and improve stand- 7 2008 Blackwell Publishing. ISBN 978-1-4051-2766-0. ards of care. The need for referral applies especially to 719 ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 720 720 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage (c) (a) (b) (f) (d) (e) Fig. 14.1 CT scans of a patient with extensive initial ischaemia. demarcated white and grey matter (black arrowheads), but This 50-year-old woman suddenly lost consciousness, and no other abnormalities that explain the coma. (d, e, f) CT remained comatose. (a, b, c) CT scan on admission showing scan 12 h later showing extensive areas of infarction subarachnoid blood in the frontal interhemispheric fissure (black arrowheads). and sylvian fissures (white arrowheads), and areas with poorly 8 patients who are designated as ‘poor grade cases’; these rehabilitation physicians, neuropsychologists, and a may be the very patients who need urgent intervention dedicated team of nurses, physiotherapists, speech because of, for example, early rebleeding, progressive therapists and occupational therapists. The neurologist brain shift from a haematoma, acute hydrocephalus or is well suited and available for assessing the patient’s hypovolaemia. For the first 2 weeks after the haemor- general medical and neurological condition on a fre- rhage, patients should be managed in an intensive or quent basis, and can act immediately if there is any medium care unit. The neurologist can play a central role clinical deterioration. Moreover, the neurologist can use in the team, which should consist of neurosurgeons, his or her experience from working in a stroke unit in interventional neuroradiologists, neuro-intensivists, the management of SAH patients. .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 721 14.2 Treatable causes of poor clinical condition on admission and their management 721 resulting in impaired or absent cerebral blood flow and 14.2 Treatable causes of poor clinical subsequent ischaemia (Fig. 14.1). condition on admission and their management 14.2.1 Early rebleeding In the first few hours after the initial haemorrhage, up to A decreased level of consciousness, with the initial 15% of patients have a sudden episode of clinical deter- haemorrhage or after early rebleeding, may be caused ioration that suggests rebleeding (Fig. 14.2). 9–11 Because by neurological complications such as an intracerebral such episodes often occur before the first CT scan, or haematoma, extensive intraventricular haemorrhage, even before admission to hospital, a definite diagnosis is 8 subdural haematoma or hydrocephalus, and by cardio- difficult and the true frequency of rebleeding on the first pulmonary complications. Only by exclusion should it day is invariably underestimated. If respiratory or cardiac be assumed that the cause is global brain damage as arrest occurs at the time of (suspected) rebleeding during a result of the high pressure after aneurysmal rupture, transportation to hospital or early after arrival at the (c) (a) (b) (d) (e) Fig. 14.2 CT scans of a patient with early rebleeding. (a,b) CT middle cerebral artery (arrow). Shortly after the initial scan the scan on admission showing large amounts of subarachnoid patient suddenly lost consciousness. (d, e) Repeated CT scan blood in the basal cisterns and left sylvian fissure. (c) CT after the clinical deterioration confirmed rebleeding with more angiogram on admission showing an aneurysm of the left extravasated blood (arrowheads). .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 722 722 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage (a) (b) Fig. 14.3 CT scans of a patient with a poor clinical condition on admission from intracerebral extension of the haemorrhage. (a) CT scan on admission showing intracerebral extension of the haemorrhage (arrowheads) and (b) shift of the lateral (arrowhead) and third ventricle (arrow) across the midline. (c) CT (c) angiogram showing an aneurysm of the right middle cerebral artery (arrow). The patient was operated upon immediately with clipping of the aneurysm and evacuation of the haematoma. (d) Repeat CT scan 3 months after the haemorrhage showing only a small loss of parenchyma (arrowhead); the patient had regained independence for activities (d) of daily living. emergency department, the patient should be resus- evacuation should be considered (Figs 14.3 and 14.4). citated and ventilated, because survival without brain The traditional intervention is evacuation of the hae- damage is still possible (section 14.6.2). 12,13 After resus- matoma with simultaneous clipping of the aneurysm citation, it will usually become clear within a matter of if it can be identified, usually only by MR angiography hours whether there is persisting brainstem dysfunction. or CT angiography rather than by catheter angiography; this course of action is backed up by a small randomized study, in which 11 of 15 patients in the operated group 14.2.2 Intracerebral extension of the haemorrhage survived, against 3 of 15 in the conservatively treated Intracerebral extension of the haemorrhage occurs in group (relative risk 0.27; 95% CI 0.09–0.74). 16 Nowadays about one-third of patients with ruptured aneurysms 14 of course this approach may be modified to immediate and, not surprisingly, the outcome is worse than in pa- occlusion of the aneurysm by endovascular coiling, tients with purely subarachnoid blood. 15 When the most followed by evacuation of the haematoma. 17 Another likely cause of deterioration in the level of consciousness option is immediate aneurysm occlusion followed by an is a large temporal lobe haemorrhage, usually from a rup- extensive hemicraniectomy that allows external expan- tured aneurysm of the middle cerebral artery, immediate sion of the brain. 18 .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 723 14.2 Treatable causes of poor clinical condition on admission and their management 723 Subdural haematomas secondary to a ruptured aneurysm may be life threatening, in which case immediate evacu- ation seems called for, despite the lack of controlled studies. 8,25 14.2.4 Acute hydrocephalus and intraventricular haemorrhage Gradual obtundation within 24 h of haemorrhage, some- times accompanied by slow pupillary responses to light and downward deviation of the eyes, is fairly character- istic of acute hydrocephalus; 26,27 and some patients lapse into coma from hydrocephalus within a few hours of the haemorrhage. If the diagnosis is confirmed by brain CT, then a lumbar puncture or early ventricular drainage may be needed, 28,29 although some patients improve spontaneously in the first 24 h. This is an area badly in need of randomized trials. Acute hydrocephalus with a large amount of intra- ventricular blood is often associated with a poor clinical condition from the outset (Fig. 14.6). In patients with ruptured aneurysms, i.e. an arterial source of bleeding, an intraventricular haemorrhage volume of 20 mL or more is invariably lethal. 30 An indirect comparison of observational studies suggests that an external ventricu- lar catheter is not very helpful, but that a strategy where Fig. 14.4 ‘Enlarging haematoma’. Top: CT scan on day of drainage is combined with fibrinolysis through the drain subarachnoid haemorrhage from a ruptured aneurysm of results in a good outcome in half the patients. 31 This the right middle cerebral artery, in a 51-year-old man on needs to be confirmed in randomized controlled trials. 32 anticoagulant treatment because of myocardial infarction with Surgical evacuation of a ‘packed’ intraventricular hae- mitral valve regurgitation. Haematoma in the right sylvian morrhage, especially in the fourth ventricle, is pointless fissure (thin arrow) and in the subinsular cortex (thick arrow); at 33 34 according to some, while others disagree. At any rate, the time of the scan anticoagulation had already been reversed ventriculostomy is certainly indicated in patients with by infusion of a concentrate of coagulation factors. Bottom: CT acute symptomatic hydrocephalus caused by a small scan 5 days after the haemorrhage, showing a marked increase of the space-occupying effect, with compression of the right clot plugging the proximal aqueduct. 35 lateral ventricle and the third ventricle. The increased mass effect results not only from a newly formed rim of oedema 14.2.5 Cardiopulmonary complications around the haematoma (short arrow), but also from an increase in size of the haematoma itself. There had been no new Pulmonary oedema and cardiac dysfunction, which can neurological signs or other clinical features indicating aggravate the pulmonary oedema, 36 may develop within rebleeding. The newly formed collection of blood in the hours of the haemorrhage, even in patients who are posterior horn of the left lateral ventricle (long arrow) is often admitted in good clinical condition (Fig. 14.7). A rapid seen in the absence of rebleeding (sedimentation effect). decline in consciousness associated with hypoxaemia and hypotension occurs in the emergency department or soon after admission. 37 According to a systematic review, 14.2.3 Acute subdural haematoma pulmonary oedema and echocardiographic abnormal- In about 2% of patients an acute subdural haematoma ities occur in 4% of patients admitted in good clinical complicates rupture of an intracranial aneurysm, usu- condition, 38 but because these complications are related ally at the origin of the posterior communicating artery to the clinical condition, in patients admitted in coma (Fig. 14.5), 19,20 more often with recurrent aneurysmal pulmonary oedema is found in one-third and echocardio- 37 rupture than a first bleed. 21 Sometimes there is no asso- graphic abnormalities in one-half. The cardiopulmonary ciated haemorrhage in the subarachnoid space to indi- dysfunction can last for weeks, but even after prolonged cate a ruptured aneurysm as the underlying cause. 22–24 intensive care treatment good recovery can still occur. 37 .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 724 724 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage Fig. 14.5 CT scans of a patient with a poor clinical condition on admission from a subdural haematoma. (a, b) Subarachnoid haemorrhage in the frontal interhemispheric fissure and the sylvian fissures with shift of the anterior interhemispheric fissure over the midline (arrow). The intracerebral extension (white open arrowhead) is too small to explain this midline shift. There is a large subdural (a) (b) haematoma (black arrowheads). (c) (a) (b) Fig. 14.6 CT scans of a patient with a poor clinical condition on admission from extensive intraventricular extension of the haemorrhage and hydrocephalus. (a) Slice showing the aneurysm (open white arrowheads), intracerebral extension of the haemorrhage (closed white arrowheads) and blood in the fourth ventricle (arrow). (b) Slice showing blood in the lateral (upper arrow) and third (lower arrow) ventricle. (c) CT angiogram showing the aneurysm on the anterior communicating artery (arrowhead). (d, e) Later CT scan showing disappearance of blood from the (d) (e) ventricles (arrows). .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 725 14.3 General care 725 (b) (a) Fig. 14.7 A 70-year-old man admitted with a subarachnoid developed shortness of breath and lapsed into coma. (a) Chest haemorrhage. On admission blood pressure was 100/50 and the X-ray on admission. (b) Chest X-ray several hours later showing pulse 50/min. He opened his eyes when spoken to, obeyed pulmonary oedema (more prominent on the right side). simple commands and was disoriented. Within hours he should not be interfered with. In keeping with this 14.3 General care notion, a further observational study from the same centre (Rotterdam) suggested that the combined strategy of avoiding antihypertensive medication and increasing fluid intake decreases the risk of cerebral infarction. 40 14.3.1 Blood pressure It seems best to reserve antihypertensive therapy for Aggressive treatment of high blood pressure carries a patients with extreme elevations of blood pressure (mean definite risk of ischaemia in brain areas with loss of auto- arterial pressure [(systolic blood pressure + 2 × diastolic regulation. The rational approach is therefore to advise blood pressure)/3] of 130 mmHg or over), and for those against treating hypertension following aneurysmal rup- with signs of rapidly progressive end organ deterioration, ture. The empirical evidence from clinical trials is sparse, such as proteinuria. An effective treatment is esmolol, but tends to support the avoidance of antihypertensive an ultra-short acting, cardioselective beta-blocking drug drugs. An observational study from the 1980s, in which (as a 0.5–1 mg/kg intravenous loading dose over 1 min, all events had been documented by serial CT scanning, followed by an infusion starting at 50 µg/kg per min and compared patients who had been treated for SAH-induced increasing up to 300 µg/kg per min as necessary), or hypertension with normotensive controls; the rate of labetalol hydrochloride, a combined α - and non-selective 1 rebleeding was lower but the rate of cerebral infarction β-adrenergic receptor blocker (as an intravenous loading was higher than in untreated controls, despite the blood dose of 20 mg, followed by repeated incremental doses pressures being, on average, higher than in the controls. 39 of 20–80 mg by intravenous bolus every 10 min until This suggests that hypertension after SAH is a compens- the desired level of blood pressure is achieved). 41 It is atory phenomenon, at least to some extent, and one that reasonable to aim for a 25% decrease in mean arterial .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 726 726 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage pressure below the initial baseline. For monitoring the of developing heart failure, but frequent calculation of blood pressure in these situations an arterial line is fluid balance (four times a day until approximately day almost mandatory. In practice, however, there is rarely 10) is the main way to estimate how much fluid should any need for antihypertensive therapy. In many patients be given to keep the patient normovolaemic. Pulmonary surges of high blood pressure can be attenuated by ade- artery catheters have been advocated on the basis of quate pain management. historical controls, 47 but in a retrospective series of 184 patients with SAH they were associated with complica- Hypertension in the acute phase of subarachnoid tions such as catheter-related sepsis (13%), congestive haemorrhage can be left untreated unless the blood heart failure (2%), subclavian venous thrombosis (1%) pressure is extremely high, or there are signs of and pneumothorax (1%). 48 Moreover, their usefulness end-organ damage. in critically ill patients in general has become uncer- tain. 49,50 The lack of evidence for prophylactic hyper- volaemia instead of aiming for normovolaemia is 14.3.2 Fluids and electrolytes discussed in section 14.5.7. Fluid management after subarachnoid haemorrhage (SAH) should aim to prevent a reduction in plasma volume, We recommend a high fluid intake of 2.5–3.5 L per because this may contribute to the development of cere- day of isotonic saline in patients with aneurysmal bral ischaemia (section 14.7). However, the arguments subarachnoid haemorrhage, to compensate for for a liberal (some might say aggressive) regimen of fluid ‘cerebral salt wasting’ and to prevent hypovolaemia administration are indirect: which predisposes to cerebral ischaemia. • Approximately one-third of SAH patients decrease their plasma volume by more than 10% between the 14.3.3 Analgesics and general nursing care second and tenth day after the haemorrhage; this is associated with a negative sodium balance; in other Headache words, there is loss of sodium as well as of water – so-called ‘cerebral salt wasting’. 42,43 Headache can sometimes be managed with just mild • Fluid restriction in patients with hyponatraemia, analgesics such as paracetamol (acetaminophen); salicy- applied in the past because it was erroneously attributed lates are best avoided because of their antihaemostatic to water retention via inappropriate secretion of anti- effect which is unwanted in patients who may have diuretic hormone, associated with an increased risk of to undergo neurosurgical clipping of the aneurysm or cerebral ischaemia. 44 external ventricular drainage; also aspirin administered • Two cohort studies with historical controls suggested after aneurysm occlusion probably does not protect that a daily intake of at least 3 L of saline (against against delayed cerebral ischaemia (section 14.5.4). In 1.5–2.0 L in the past) was associated with a lower risk practice, the pain is usually so severe that codeine needs of delayed cerebral ischaemia and a better overall out- to be added, which will not mask neurological signs. come. 40,45 The interpretation of these two studies is Sometimes, pain can be alleviated with anxiolytic drugs difficult not only because of their observational nature, such as midazolam (5 mg via an infusion pump). Often a but also because the liberal administration of saline in synthetic opiate such as tramadol is needed to obtain the second period was accompanied by avoidance of relief (for doses see Table 14.1). As a last resort, severe antihypertensive drugs as well. headache can be treated with piritramide. Constipation Despite the incomplete evidence, it seems reasonable is a common problem with all opiates. to prevent hypovolaemia. We aim for normovolaemia by giving 2.5–3.5 L/day of isotonic saline, unless con- In patients with a decreased level of consciousness traindicated by signs of impending cardiac failure (see headache may present as restlessness. The ‘analgesic below). The amount of intravenous saline should be staircase’ consists of frequent doses of paracetamol, reduced if the patient is also receiving nutritional solu- then codeine if necessary, tramadol or, as a last resort, tion via the enteral route. In patients with fever, from piritramide. whatever cause (section 14.3.3), fluid intake should be gradually increased. 46 For as long as the aneurysm has not been occluded the Fluid requirements may be guided by recording the patient should be kept on complete bed rest, tradition- central venous pressure (the directly measured value ally flat and with as few external stimuli as possible, on should be above 8 mmHg), especially in patients with the assumption that any form of excitement increases hyponatraemia or a negative fluid balance who are at risk the risk of rebleeding. .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 727 14.3 General care 727 Table 14.1 Recommendations for Nursing nursing and general management of Continuous observation (Glasgow Coma Scale, temperature, ECG monitoring, patients with subarachnoid haemorrhage. pupils, any focal deficits) Nutrition Oral route Only with intact cough and swallowing reflexes Keep stools soft by adequate fluid intake and by restriction of milk content; if necessary add laxatives Nasogastric tube Deflate endotracheal cuff (if present) on insertion Confirm proper placement by X-ray Begin with small test feeds of 5% dextrose Prevent aspiration by feeding in sitting position and by checking gastric residue every hour Tablets should be crushed and flushed down Total parenteral nutrition should only be used as a last resort Blood pressure Do not treat hypertension unless the mean arterial pressure is ≥ 130 mmHg or there is clinical or laboratory evidence of progressive end organ damage Fluids and electrolytes Intravenous line mandatory Give at least 3 L/day (isotonic saline, 0.9%) Insert an indwelling bladder catheter Compensate for a negative fluid balance and for fever Monitor electrolytes (and white blood cell count) at least every other day Pain Start with paracetamol (acetaminophen) 500 mg every 3–4 h; avoid aspirin Midazolam can be used if pain is accompanied by anxiety (5 mg via infusion pump) For severe pain, use codeine (30–60 mg every 4 h, as needed), tramadol (50–100 mg every 4 h, as needed) or, as a last resort, piritramide 0.2–0.3 mg/kg body weight i.m. (maximum 80 mg per 24 h, in four doses). Prevention of deep venous thrombosis and pulmonary embolism Compression stockings or intermittent compression by pneumatic devices, or both before aneurysm occlusion. Low molecular heparin if the aneurysm has been treated. Stool softeners should be prescribed routinely. Enemas Monitoring conscious level are contraindicated because they markedly increase intra- Continuous assessment of the consciousness level is abdominal pressure and secondarily intracranial pressure. essential, preferably recorded with the Glasgow Coma Scale (section 3.3.2), although it should be remem- The bladder bered that with inexperienced observers errors of more than one point may occur. 51,52 Any change in the level A catheter is almost always needed for accurate calculation of consciousness may signify early cerebral ischaemia, of fluid balance, except in men who are alert and have rebleeding, hydrocephalus or a systemic medical com- perfect bladder control. Condom devices leak or slip off plication. The nurses should be familiar with the peak too often to be of any use in these critically ill patients. times of occurrence of rebleeding and cerebral ischaemia, Intermittent catheterization may substantially decrease and they should be able to detect the early signs of the risk of urinary infections but the procedure is too delirium, which is seldom caused by SAH per se but more stressful for patients with SAH (and requires more time often by the combined effects of isolation, sleep depriva- from the nursing staff than most hospital budgets allow). tion, narcotic drugs and alcohol withdrawal. Confusion Constipation In confused patients specific causes of confusion should Coughing and straining must be rigorously prevented be excluded, such as pain, hypoxia, alcohol withdrawal, because of the attendant surges in arterial blood pressure. infection or a blocked urinary catheter. .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 728 728 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage legs; a study specifically performed in patients with SAH Venous access but with non-randomized controls suggested DVT can be A reliable intravenous route should be maintained for successfully prevented in this way. 58 And in a random- emergency administration of plasma expanders or anti- ized trial, but in patients with intracerebral haemorrhage, epileptic drugs. Oral administration of antacids is said to the combination of stockings and intermittent pneu- be sufficient to protect against stress bleeding; proton matic compression resulted in a smaller risk of DVT than pump inhibitors should be considered only in mechan- prevention by stockings alone (relative risk 0.29; 95% CI ically ventilated patients or in patients with a previous 0.08–1.00); there is no reason to assume that this effect 59 history of ulcers. would be any different in patients with SAH. The devices are well tolerated by patients and can be managed easily by the nursing staff but one problem is that their success Fever depends on a repair service that keeps them running. The body temperature should be measured frequently, It has been argued that impedance plethysmography up to four times a day, depending on the time since the or duplex Doppler sonography should be performed to SAH, and the level of consciousness. After the first few exclude clinically occult thrombosis before any kind of hours mild fever (up to 38.5°C) often occurs, probably as compression is applied to the calves, but no systematic the result of the inflammatory reaction in the subarach- studies have answered the question whether the danger noid space, 53 in which case the pulse is characteristically of dislodging emboli to the lungs is imaginary or not. normal. 54 Infection should be suspected if the temper- ature exceeds 38.5°C, if the pulse is elevated as well, or Antiepilpetic drugs if the patient has vomited. An elevated white cell count is not helpful in distinguishing infection from non- The prophylactic use of anti-epileptic drugs is controver- infective causes of pyrexia. 46 sial. Seizures in the first few weeks after aneurysmal rup- ture occur in about 10% (6–16%) of patients, and may be easily confused with rebleeding. 60–62 The majority occur Venous thromboemolism 63 soon after the initial haemorrhage. Predictive factors Deep venous thrombosis (DVT) is not as common after for epileptic seizures have not been consistently iden- SAH as in patients with ischaemic stroke, presumably tified. Intracranial surgery, 64 and probably also intra- because the patients are restless, generally younger and, cerebral extension of the haemorrhage, increase the most important, seldom have a paralysed leg; in a large risk, but a randomized trial of anti-epileptic drugs after and prospective series DVT was clinically diagnosed in supratentorial craniotomy for benign lesions (not only only 4%. 55 DVT can be prevented by subcutaneous low- aneurysms) in 276 patients failed to show benefit in dose heparin or heparinoids (section 11.13), but the terms of seizure rate or case fatality, although the con- obvious fear is that anticoagulation will not be confined fidence intervals were wide. 65 Currently, we adhere to to the venous system. In a placebo-controlled trial of the principle of ‘in dubio abstine’ (abstain when in doubt) the low-molecular-weight heparinoid enoxaparin given because the possible disadvantage of a serious drug reac- after aneurysm occlusion, intracranial bleeding com- tion may well outweigh any benefits. plications occurred more often in the treatment group, Our recommendations for the general management of while there was no influence on overall outcome or the patients with SAH are summarized in Table 14.1. Before risk of delayed ischaemia. 56 In the absence of proof of considering the diagnosis and prevention of all the effectiveness, many physicians prefer to err on the side possible complications we will discuss the prevention of of caution and do not recommend prophylactic heparin the two most common causes of clinical deterioration: or the low-molecular-weight heparinoids. rebleeding and cerebral ischaemia. Graduated compression stockings reduce the risk of DVT in patients undergoing general surgical, neurosurgical or orthopaedic procedures (section 11.13). 57 There- fore, because low-molecular-weight heparinoids increase the risk of intracranial bleeding, 56 graduated compres- 14.4 Prevention of rebleeding sion stockings are the preferred form of DVT prevention in SAH patients – although admittedly this advice lacks support from a randomized trial in this specific situation. 14.4.1 Risk of rebleeding However, because compression stockings must be indi- vidually fitted to be efficacious, some favour pneumatic In the first few hours after admission for the initial devices that apply intermittent venous compression to the haemorrhage, up to 15% of patients experience a sudden .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 729 14.4 Prevention of rebleeding 729 episode of clinical deterioration that suggests rebleeding Rebleeding of a ruptured aneurysm cannot be (section 14.2.1). In patients who survive the first day, the predicted reliably. risk of rebleeding is more or less evenly distributed over the next 4 weeks. 66 After the first day, the cumulative risk 14.4.2 Endovascular occlusion within the first 3 weeks, and without measures to pre- vent it, is about 40%. 67 Between 4 weeks and 6 months Over the last decade, endovascular occlusion of aneurysms after the haemorrhage the risk of rebleeding gradually has largely replaced surgical occlusion as the interven- declines, from the initial level of 1–2%/day to a constant tion of choice for the prevention of rebleeding, at least in level of about 3%/year 68 (section 14.10.2). centres where this expertise is available. The technique consists of packing the aneurysm with platinum coils, The risk of rebleeding after rupture of an intracranial nowadays with a system for controlled detachment. 76 aneurysm, without medical, endovascular or surgical This was initially mainly used for posterior circulation intervention, is about 40% in the first 3 weeks, aneurysms (Fig. 14.8), and then later for aneurysms on i.e. 1–2% per day. the anterior circulation (Figs 14.9 and 14.10). The pro- cedure is generally performed under general anaesthesia, After rebleeding the prognosis is poor; 80% of patients though not invariably. 77 4 die or remain disabled. Indeed, given its frequency and Remarkably soon after its introduction in the early morbidity, rebleeding is still a major cause of poor out- 1990s coiling was compared with neurosurgical clip- come, even in centres aiming at early occlusion of the ping in randomized trials. A systematic review of these aneurysm. 45,69,70 Unfortunately, there are few prognostic trials (one large and two small trials, of which one was factors that predict an increased risk of rebleeding. Some still unpublished) included a total of 2272 patients studies suggest that it occurs more often in patients with (Fig. 14.11). 78 Most of the patients were in good clinical a previous bout of sudden headache, 71 with a decreased condition and had an aneurysm of the anterior circula- level of consciousness, 72 or with larger aneurysms 72–74 tion. After 1 year of follow-up, the relative risk of a poor but none of these factors reliably identify the patients outcome (death or dependency) for coiling vs clipping at highest risk. And even if high-risk groups could be was 0.76 (95% CI 0.67–0.88). The absolute risk reduction identified, there would still be almost as many rebleeds of a poor outcome was 7% (95% CI 4–11%). For anterior because most occur in the larger number of patients at circulation aneurysms the relative risk of a poor outcome moderate risk – the prevention paradox again (section was 0.78 (95% CI 0.68–0.90) and the absolute risk reduc- 18.5.1). 75 tion was 7% (95% CI 3–10%). For posterior circulation (a) (b) Fig. 14.8 (a) Catheter angiogram showing aneurysm at the tip aneurysm has been occluded by detachable coils, through an of the basilar artery (arrow). (b) Repeat angiogram after the endovascular approach. .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 730 730 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage Several qualifications are needed to avoid the impres- sion that these results fully define the relative merits of endovascular and surgical occlusion of aneurysms: • First, aneurysms in some locations are more suitable for occlusion by one rather than the other technique. Basilar artery aneurysms and many other types of posterior circulation aneurysms are relatively easy to coil, 80 whereas surgical treatment is often very diffi- cult and complicated. Conversely, aneurysms at the trifurcation of the middle cerebral artery are often dif- ficult to coil without interfering with major arterial branches. These preferences are reflected in the under- representation of posterior circulation aneurysms and aneurysms of the middle cerebral artery among the 2143 patients randomized in by far the largest trial, the international ISAT study. 81 • Second, patients over 70 years old were under- represented in the ISAT trial, reflecting a preference for coiling – or perhaps for no treatment – in many aged patients with SAH. However, it is not an unrealistic (a) guess that the relative advantages of coiling over sur- gical treatment are even greater in patients over the age of 70 than in younger patients, given that the risk of complications from treatment of unruptured aneurysms in older patients is much higher for opera- 5 tion than for coiling. SAH patients older than 70 or 75 years of age have in general a poor prognosis, but several observational studies suggest that they can still benefit from endovascular or surgical occlusion of the aneurysm, especially if admitted in a good condi- tion; without treatment the most important cause of a poor outcome is recurrent haemorrhage. 82 • A third and last concern is the durability of the occlu- sion with coils, given that the early rebleeding rate after surgical occlusion is very low (section 14.10.2). Ideally, after coiling, any remaining aneursymal lumen becomes occluded by a process of reactive thrombosis, but early or late rebleeding does nevertheless occur – even after a technically competent procedure. In a consecutive series of 431 patients, six (1.4%) rebled within 30 days after coiling; in four of these patients the aneurysm (b) 83 had been completely occluded during the procedure. Fig. 14.9 Endovascular occlusion of an aneurysm on the All six patients died after rebleeding. Risk factors for anterior communicating artery. Same patient as in Fig. 14.6. rebleeding in this series were a haematoma in the (a) Catheter angiogram before occlusion showing the aneurysm vicinity of the aneurysm, anterior communicating (arrow) and branching vessels (arrowheads). (b) Catheter aneurysms, and small aneurysm size. In the ISAT trial angiogram after occlusion showing the coils mass in the the rate of rebleeding from the target aneurysm aneurysm (arrow) and the still patent branching vessels between the procedure and the end of the first year (arrowheads). was somewhat higher after coiling than clipping (2.6% aneurysms the relative risk was 0.41 (95% CI 0.19–0.92) vs 1.0%), but the initial 7.4% absolute gain in the and the absolute risk reduction 27% (95% CI 6–48%). avoidance of death or dependence was maintained at Also the cognitive impairment after 1 year tends to be 1 year, while the early survival advantage was still seen less after coiling than after clipping. 79 to 7 years. 64 Most episodes of rebleeding occurred in .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 731 14.4 Prevention of rebleeding 731 (b) (a) (c) (d) (e) Fig. 14.10 Endovascular occlusion of an aneurysm on the left (d) MRA after 6 months suggests some reopening of the middle cerebral artery. (a) Catheter angiogram before occlusion neck of the aneurysm (arrow). (e) Catheter angiogram after showing the aneurysm (arrow). (b) Three-dimensional 6 months confirming the reopening of the neck (arrow). The reconstruction showing detailed delineation of the aneurysm fundus of the aneurysm is still occluded by the coils mass with relatively wide neck (arrow). (c) Catheter angiogram after (arrowhead). On further follow-up angiograms the reopening occlusion showing the coils mass in the aneurysm (arrow). remained the same and no further treatment was performed. Study coil clip RR (fixed) Weight RR (fixed) or sub-category n/N n/N 95% CI (%) 95% CI Brilstra 2000b 3/8 4/8 1.16 0.75 [0.24, 2.33] Koivisto 2000 11/52 14/57 3.88 0.86 [0.43, 1.72] ISAT 250/1063 326/1055 94.96 0.76 [0.66, 0.88] Total (95% CI) 1123 1120 100.00 0.76 [0.67, 0.88] Total events: 264 (coil), 344 (clip) 2 2 Test for heterogeneity Chi = 0.12, df = 2 (P = 0.94), I = 0% Test for overall effect: Z = 3.83 (P = 0.0001) 0.1 0.2 0.5 1 2 5 10 Favours coiling Favours clipping Fig. 14.11 Forest plot of Cochrane review of endovascular coiling vs surgical clipping in patients with aneurysmal subarachnoid haemorrhage (SAH). Outcome, death or dependency at 12 months after SAH. RR, relative risk. .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 732 732 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage the first 30 days, with either procedure. Rebleeding 3 months of therapy with antiplatelet drugs but this is after the first month occurred in five of a cohort of 381 supported only by comparisons with historical con- patients (1.3%) with a mean period of follow-up of trols, 94 not by randomized trials. 95 4 years. 84 More data on the risk of late rebleeding Another complication is that one or more coils after coiling will follow in the near future, but in any may become dislodged and herniate or embolize into event this must be weighed against the risk of late the parent vessel, which again may lead to ischaemic rebleeding after surgical treatment, which is around brain damage, unless a surgical approach retrieves the 3% after 10 years. 85,86 In a multicentre study that com- fragment. 96,97 pared rebleeding after the initial year after coiling Finally, rare complications include secondary infec- or clipping, rebleeding rates were low in both cohorts tion of the thrombosed aneurysm, 98 worsening of mass 99 after a relatively short period of follow-up (mean effect, intracranial artery dissection, and false aneurysms 4.4 years). 87 Rebleeding after 1 year occurred in one of the femoral artery. patient treated with coil embolization during 904 In some patients with large or complex aneurysms person-years of follow-up (annual rate 0.11%; 95% coiling is combined with surgical treatment, 100,101 or CI 0–0.63%) and in no patients treated with surgical with stenting. 102 Stents may also be useful in fusiform clipping during 2666 person-years (annual rate 0%; aneurysms. 103,104 95% CI 0–0.14%). Follow-up When to coil There is still uncertainty about how long patients need The optimal timing of coiling has not been studied, to be followed up with imaging beyond the period of but given the risk of rebleeding in the first few days, the 6 months that is fairly standard, and also about the most general opinion is ‘the sooner the better’. 88 If endo- suitable radiological technique. Conventional catheter vascular occlusion is not possible in the first few days, angiography is time-consuming and carries a small there is probably no need to avoid the period between but definite risk. MR angiography is feasible and gives day 4 and 12 which is regarded as unfavourable for good-quality images, 105–108 but its test characteristics surgical intervention (section 14.4.3), since delayed cere- and effectiveness have not yet been studied in large bral ischaemia is relatively unlikely to be precipitated by series of patients. endovascular occlusion. 89 A final area of uncertainty is the need for a second procedure for aneurysm necks that have recanalized by impaction of the coils, by re-coiling 109 or surgical occlu- The complications of coiling 110 sion. Since rebleeding can occur after incomplete Bearing in mind that coiling is not a technically feas- aneurysm occlusion and re-coiling has a low risk of ible option in every patient, 90 a problem that can be complications, 109 currently most teams recommend reduced by the use of three-dimensional angiography, 91 re-coiling, but the risk : benefit ratio of this strategy has and despite its many advantages over surgical treatment, not yet been properly assessed. coiling is not without its hazards. In a large consecutive New developments are radioactive platinum coils and series from a single centre, procedural complications hydrogel-coated coils, 111,112 but their usefulness has not occurred in 40 of 681 patients (5.9%; 95% CI 4.2–7.9%), yet been properly evaluated. leading to death in 18 (procedural mortality, 2.6%; 95% CI 1.6–4.2%) and to disability in another 22 patients 14.4.3 Surgical occlusion (procedural morbidity, 3.2%; 95% CI 2.0–4.9%). 92 The most frequent complication was thromboembolism, Surgical occlusion of ruptured aneurysms has now which occurred in 4.7% of patients; intraprocedural become the second choice, at least for most patients re-rupture occurred in 1.2% of patients (Fig. 14.12). If and most aneurysms. A statistical modelling exercise rebleeding occurs during coiling the bleeding is much that weighed the prevention of rebleeding against the more difficult to control than during surgical explora- complications of the operation at current standards tion, which explains the high case fatality of 40%. 93 Risk estimated an absolute reduction in the risk of a poor factors for procedural complications are small aneurysm outcome of almost 10% and a relative risk reduction size and the use of a temporary occlusion balloon. 92,93 To of 0.81 for surgical occlusion vs no occlusion. 67 In fact, reduce the risk of thromboembolism, most interven- no proper randomized trial has ever been done in tional radiologists start heparin during treatment and modern times; the only randomized trial that compared advise continuing for at least 1–2 days. Some advise even surgical with conservative treatment in SAH patients, .. ..
9781405127660_4_014.qxd 10/13/07 2:14 PM Page 733 14.4 Prevention of rebleeding 733 (c) (b) (a) (d) (e) (f) Fig. 14.12 CT scans and angiograms of a patient with rupture of the aneurysm during coiling. (a) CT on admission showing blood in the basal cisterns and fissures (arrowheads), and in the fourth ventricle (arrow). (b) Reconstruction of the CT angiogram showing an aneurysm (circle) at the origin of the posterior inferior cerebellar artery (which in this patient branches at the junction of the vertebral and basilar artery). (Also reproduced in colour in the plate section.) (c) Three-dimensional reconstruction of the catheter angiogram showing detailed delineation of the aneurysm. (d) Catheter angiogram before the coiling procedure showing the aneurysm (arrow). (e) Catheter angiogram during the coiling procedure showing the partially coiled aneurysm (arrow) and extravasation of contrast (arrowheads). (f) Catheter angiogram after occlusion showing the coils mass in the aneurysm (arrow). (g) CT scan after the procedure showing contrast material in the basal cisterns (arrowheads) and fourth ventricle (arrow). (g) .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 734 734 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage performed in the 1950s in London, tested carotid ligation, Although early aneurysm operation is now the not aneurysm clipping, and was restricted to patients usual practice when surgical clipping is indicated, who had survived the first 12 days. 113,114 this policy is vulnerable to challenge and reversal Nowadays, aneurysms are usually clipped early, i.e. because it has not been supported by any randomized within 3 days of the initial bleed – if possible within controlled trial. 24 h – and this policy is no longer restricted to patients admitted in good clinical condition. 70 The main ratio- Neurosurgical and anaesthetic techniques have under- nale for early surgery is, of course, optimal prevention gone considerable evolution other than through the of early rebleeding. The only randomized trial of the tim- introduction of the operating microscope. For example, ing of operation, performed in Finland, allocated 216 three-dimensional angiography facilitates accurate patients to one of three groups: operation within 3 days, placing of the occlusive clip. 120 Mild intra-operative after 7 days, or in the intermediate period. 115 The out- hypothermia confers no benefit. 121 Although this is not come tended to be better after early than intermediate or a textbook about technical details of operative treatment, late surgery, but as the difference was not statistically it is useful for physicians to know something about significant a disadvantage of early surgery could not be these developments and Table 14.2 lists some of them. excluded. 115,116 The same result – no difference in out- A most unusual complication of operative treatment of come after early or late operation – also emerged from aneurysms is a delayed type hypersensitivity against the observational studies. 117–119 metal alloy presenting as intense generalized pruritus, Some believe that early operation also helps prevent which may necessitate removal of the clip. 122 ischaemia, because the clots that surround blood vessels It should not be assumed that surgical treatment is in the subarachnoid space can be washed away and no always definitive; rebleeding from the treated aneurysm longer contribute their assumed effect in the develop- can occur even in the first year after rupture, and also ment of vasospasm, but there is little evidence to support later during follow-up at which time haemorrhages from this notion (section 14.5.8). Another theoretical advant- newly developed aneurysms at the site of the original age of early aneurysm clipping is that if cerebral ischaemia aneurysm or at other sites, or from small unruptured does develop hypertensive and hypervolaemic treat- incidental aneurysms that were left untreated at time of ment can be given without the danger of rebleeding the subarachnoid haemorrhage, may begin to be seen (section 14.7.2). (section 14.10.2). 123 Table 14.2 Special neurosurgical Vessel reconstruction, with fusiform aneurysms or giant aneurysms techniques used for occlusion of Permanent occlusion of parent vessel 319–321 aneurysms. Temporary occlusion of parent vessel 322,323 ‘Trapping’ without revascularization 320 Opening the aneurysm for thrombectomy, followed by reconstruction of the artery 319,324 Excision of aneurysm, with end-to-end anastomosis of parent artery 319,325 Excision of aneurysm, with large-artery bypass, supratentorially, 326,327 or infratentorially 321,328 Combination with endovascular techniques Intra-operative angiography, to verify clip placement 329,330 Incomplete clipping, followed by endovascular coiling 331,332 Combination with endoscopy, to verify clip placement 333 Combination with intra-operative MR angiography, to verify clip placement 334 Special clips Clips with different shapes and physical properties 335 Titanium clips, which do not cause artifacts on MR imaging 336 Approaches to aneurysms in specific locations Distal part of posterior inferior cerebellar artery: skull-base and far-lateral transcondylar approach 321 Non-branching site of intracranial internal carotid artery (‘blister-like aneurysms’): clipping or wrapping, 337 or suturing followed by an encircling clip, 338 or balloon occlusion followed by trapping 339 Middle cerebral artery: side-to-side anastomosis after unplanned occlusion of a major branch 340 .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 735 14.5 Prevention of delayed cerebral ischaemia 735 But this is not necessarily the end of the story if measures 14.4.4 Pharmacological treatment could be found to prevent the attendant ischaemia. Antifibrinolytic drugs Recombinant factor VIIa Because fibrinolytic activity in the cerebrospinal fluid is thought to break down clots sealing the tear in the Theoretically, this activated coagulation factor might aneurysm sac, it is logical to suppose that rebleeding prevent rebleeding. Unfortunately, although an open- might be prevented by antifibrinolytic drugs which cross label, dose escalation safety study showed no evidence the blood–brain barrier rapidly after SAH. 124 The two of ischaemic complications in the first nine patients, most commonly used are tranexamic acid (TEA; 1 g intra- it was suspended when the tenth patient developed venously or 1.5 g orally, four to six times daily), or epsilon- middle cerebral artery branch occlusion contralateral to aminocaproic acid (EACA; 3–4.5 g 3-hourly intravenously the aneurysm. 127 or orally). Both agents are structurally similar to lysine and so block the lysine binding sites by which the plas- minogen molecules bind to complementary sites on fibrin. It is assumed that it takes 36 h to achieve complete inhibition of fibrinolysis in the cerebrospinal fluid. 14.5 Prevention of delayed cerebral Ten randomized trials of antifibrinolytic drugs in SAH ischaemia have so far been performed. The most recent Cochrane review included nine of them, involving 1399 patients. Treatment did not significantly influence death from 14.5.1 Pathogenesis and risk factors all causes (OR 0.99; 95% CI 0.79–1.24) or poor outcome (death, vegetative state or severe disability; OR 1.12; 95% Unlike ischaemic stroke as a result of disease of the extra- CI 0.88–1.43). On the other hand, there were striking or intracranial arteries, cerebral ischaemia or infarction differences in specific event rates: antifibrinolytic treat- after subarachnoid haemorrhage (SAH) is not usually ment reduced the risk of rebleeding at the end of follow- confined to the territory of a single cerebral artery or one up, with some heterogeneity between the trials (OR 0.55; of its branches. 128 Vasospasm is often implicated as the 95% CI 0.42–0.71), but increased the risk of cerebral cause because its peak frequency from day 5 to 14 coin- ischaemia (OR 1.39; 95% CI 1.07–1.82) with consider- cides with that of delayed ischaemia, and also because able heterogeneity between the most recent study, in it is often generalized, in keeping with the multifocal which specific treatments to prevent cerebral ischemia or diffuse nature of the clinical manifestations and of were used, 125 and the four older studies. However, even the ischaemic lesions found on brain CT and at autopsy. in the trial where antifibrinolytic agents were tested in Nevertheless, we have strong objections to the wide- the presence of nimodipine and a strategy avoiding spread use of the term ‘vasospasm’ as a synonym for hypovolaemia, the drug still did not improve overall out- delayed cerebral ischaemia, not so much out of fast- come. 125 The tenth and most recent trial was performed idiousness on our part, but because the exchange of one in Sweden and included 505 patients; but again the over- term for the other impedes understanding and, even all outcome did not appreciably improve in patients worse, progress. The eventual goal of treatment in this treated with TEA, despite an impressive reduction in area is preventing cerebral ischaemia, not preventing rebleeding. 126 arterial narrowing. There are at least five reasons why vasospasm is not Antifibrinolytic drugs prevent rebleeding after synonymous with delayed cerebral ischaemia: aneurysmal rupture, but because they increase the • First, arterial narrowing does not necessarily signify risk of cerebral ischaemia they have no useful effect contraction of smooth muscle in the arterial wall, but on overall outcome. may represent necrosis and secondary oedema, 129–131 or intimal proliferation. 132,133 How antifibrinolytic treatment precipitates cerebral • Second, arterial narrowing can be asymptomatic even ischaemia is still unclear. Possible explanations include if severe. 134–136 increased blood viscosity, the development of micro- • Third, interpretation of the severity or even the pres- thrombi, delayed clearance of blood clots around the ence of arterial narrowing on an angiogram is subject arteries at the base of the brain, and the development to considerable variation between observers. 137 of hydrocephalus through delayed resorption of blood. • Fourth, strategies that reduce vasospasm may not be In brief, antifibrinolytic drugs work, but they do not help. beneficial (endothelin receptor antagonists; section .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 736 736 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage 14.5.6) and strategies that reduce the risk of cerebral intermediate factors, and these interactions are complex ischaemia may do so without influencing vasospasm and poorly understood. Also one should bear in mind (calcium antagonists; section 14.5.2). that a sizeable proportion of patients who eventually • Last, arterial narrowing is only one of the many factors develop delayed ischaemia have none of these risk involved in the pathogenesis of cerebral ischaemia factors. 138 An often quoted study postulated a close rela- in patients with ruptured aneurysms. One-third of tionship between the location of subarachnoid blood patients with vasospasm do not develop cerebral and the ‘thickness’ of the clot on the one hand with the ischaemia and one-third of patients with cerebral occurrence of vasospasm and delayed cerebral ischaemia ischaemia have no vasospasm (Fig. 14.13). 138 on the other, but these observations were based on only Many of the predictive factors for delayed cerebral 41 patients. 139 Moreover, the method of assessing local ischaemia (Table 14.3) are interrelated or depend on amounts of subarachnoid blood (called the Fisher scale, (c) (a) (b) (f) (d) (e) Fig. 14.13 CT scans and CT angiograms of a patient who admission with normal size of intracranial vessels and an deteriorated 5 days after the onset of subarachnoid aneurysm of the anterior communicating artery (arrow). haemorrhage. The cause of the deterioration was secondary (d, e) Repeat CT scan after the clinical deterioration showing ischaemia; CT angiograms show unchanged diameter of the areas of infarction in the territory of the right anterior cerebral intracranial vessels of the circle of Willis. (a, b) CT scan on artery (black arrowheads). (f) CT angiogram after the admission showing diffuse subarachnoid blood in the basal deterioration shows unchanged diameter of the intracranial cisterns and fissures (white arrowheads). (c) CT angiogram on vessels. An intraventricular drain has been inserted (arrow). .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 737 14.5 Prevention of delayed cerebral ischaemia 737 Table 14.3 Factors at baseline and during the clinical course perfusion deficit at the time of bleeding is an independ- that have been associated with delayed cerebral ischaemia after ent risk factor for the development of delayed cerebral subarachnoid haemorrhage. ischaemia. 147 At baseline: Cerebral ischaemia after subarachnoid haemorrhage Cerebral arterial narrowing on initial angiogram 139,341 occurs only if the source is a ruptured aneurysm; but also: Previous cigarette smoking? 342 whether or not it develops is strongly related to the Previous cocaine use 343 total amount of subarachnoid blood, much less to Loss of consciousness > 1 h at onset 142,148 the distribution of the extravasated blood. Amount of subarachnoid blood on early CT scanning 142,344–347 The search for chemicals that might be mediators Location of ruptured aneurysm on the anterior for arterial narrowing and the development of delayed communicating artery or internal carotid artery, 346 or on cerebral ischaemia has been intensified as several new the anterior choroidal artery 348 candidate molecules have been recognized (Table 14.4). Acute hydrocephalus 26,143,193,349 However, the multitude of these isolated and often Perfusion deficit on initial CT scan 147 conflicting findings attests to the paltriness of our cur- 4G-allele in the 4G/5G-promotor polymorphism in the plasminogen activator inhibitor-1 (PAI-1) gene 350 rent insights into the chain of molecular events between aneurysmal rupture and delayed cerebral ischaemia. At During the clinical course: Hyponatraemia and hypovolaemia 42–44 any rate, the relationship between the amount of extra- Treatment with antihypertensive drugs 39 vasated blood and the development of delayed cerebral Hypotension during anaesthesia 351 ischaemia is not sufficiently explained by the simplistic Treatment with antifibrinolytic drugs 352 notion that toxic substances are released from clots Emboli on transcranial Doppler monitoring after surgical around the large arteries at the base of the brain. Because clipping, 353 but not confirmed in another study 354 loss of consciousness at the time of the haemorrhage Early ischaemic lesions on MR scanning 355 is an important and independent predictive factor for Impaired cerebral autoregulation 136,356 Table 14.4 Intermediate substances implicated in the after the last author), has wide inter-observer varia- pathogenesis of delayed cerebral ischaemia. For many, the tion. 140 Not surprisingly, therefore, in a comparison with relation with delayed cerebral ischaemia could not be confirmed in other studies. a method of assessing the amount of blood in each of the cisterns and fissures separately in a semi-quantitative 357 nitric oxide manner, this Fisher scale performed worse: it did not endothelin 358–362 predict delayed cerebral ischaemia and poor clinical protein C kinase 358 outcome, whereas the semi-quantitative scale did. 141 lipid peroxides 363 Although there is little doubt that the overall amount free fatty acids 364 of subarachnoid blood as determined on CT scan is a soluble intercellular adhesion molecule-1 (ICAM-1) 365,366 powerful predictor of delayed cerebral ischaemia, larger P-selectin and L-selectin 367 368 series have failed to show any relationship between sP-selectin 369 the anatomical distribution or even the side of infarc- E-selectin von Willebrand factor 370,371 tion and the site of the extravasated blood in the sub- D-dimer 349,372 arachnoid space, 128,142 or at best in half the patients. 143 prothrombin fragments I and II, thrombin-antithrombin III Another awkward fact – often conveniently overlooked 372 complex – is that delayed cerebral ischaemia is rare after SAH plasminogen activator inhibitor-1 372,373 from arteriovenous malformations despite arterial nar- vascular endothelial growth factor 370 rowing in some cases, 144 and also after bleeding from inflammatory cytokines 374,375 traumatic brain injury, 145 while it does not occur at all matrix metalloproteinase-9 370 in perimesencephalic non-aneurysmal SAH, even after brain natriuretic peptide 376 matching for the amount of extravasated blood. 146 It transferrin 377 seems that subarachnoid blood is not a sufficient factor adrenomedullin (protective) 378 379,380 for the development of delayed cerebral ischaemia – tissue oxygen pressure lactate in CSF 381,382 the source has to be a ruptured artery. This suggestion is excitatory amino acids 382 further supported by the finding that a generalized .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 738 738 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage (c) (a) (b) (d) (e) (f) Fig. 14.14 CT scans of a patient with subarachnoid haemorrhage from an anterior communicating artery aneurysm with long-lasting loss of consciousness after the haemorrhage. During the following days his clinical condition improved, but on the sixth day his level of consciousness decreased again. (a, b, c) CT on admission showing intracerebral extension of the haemorrhage (arrows) and blood in the left sylvian fissure (arrowhead). There are already some areas of hypodensity (black arrowheads). (d) CT angiogram on admission showing an aneurysm on the anterior communicating artery (white open arrow). The left A1 part of the anterior communicating artery is present (black arrows), but on the right side this artery is missing (arrow). (e, f, g) CT scans after the clinical deterioration showing large areas of infarction in the territories (g) of both anterior cerebral arteries (black arrowheads). delayed cerebral ischaemia, 142,148 it is conceivable that complications, such as arterial narrowing and hypo- global ischaemia during this brief period, along with a volaemia (Fig. 14.14). massive increase in intracranial pressure, may sensitize Practical measures that may help to prevent ischaemia neurones to marginal perfusion associated with later are first of all avoidance of antihypertensive drugs .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 739 14.5 Prevention of delayed cerebral ischaemia 739 (section 14.3.1) and an adequate intake of fluid and If the patient is unable to swallow, the nimodipine sodium, although the evidence for these recommenda- tablets should be crushed and washed down a nasogastric tions is rather slender (section 14.3.2). Other interven- tube with normal saline. Intravenous administration is tions will be discussed below. advocated by the manufacturer and is more expensive, but there is no evidence to support this. 149 Moreover, intravenous administration of nicardipine does not 14.5.2 Calcium antagonists 149 improve outcome. The lack of effectiveness of cal- A Cochrane review has assessed the trials of calcium cium antagonists via the intravenous route is probably antagonists in patients with aneurysmal SAH: 149 nimo- explained by the resulting hypotension, 153 which may dipine (eight trials, 1574 patients), nicardipine (two even be a problem if nimodipine is given orally. There- trials, 954 patients), AT877 (one trial, 276 patients) fore, if no other cause for hypotension is found, the and magnesium (one trial, 40 patients). Overall, calcium dose of nimodipine should be at first halved (to 30 mg antagonists reduced the risk of poor outcome: relative every 4 h) and subsequently discontinued if the blood risk (RR) 0.82 (95% CI 0.72–0.93); the absolute risk pressure continues to fall or the patient remains reduction was 5.1%, and the corresponding number of hypotensive. patients needed to treat to prevent a single poor outcome event was 20. For oral nimodipine alone the RR was 0.70 14.5.3 Magnesium sulphate (0.58–0.84). The RR of death on treatment with calcium antagonists was 0.90 (95% CI 0.76–1.07), of clinical signs Hypomagnesaemia occurs in more than 50% of pa- of delayed cerebral ischaemia 0.67 (95% CI 0.60–0.76), tients with SAH and is associated with delayed cerebral and of CT or MR confirmed infarction 0.80 (95% ischaemia and poor outcome. 154 Magnesium reduces CI 0.71–0.89). In brief, the risk reduction for ‘poor out- infarct volume after experimental SAH in rats 155 and its come’ is statistically robust, but depends mainly on trials putative modes of action are inhibition of the release of with oral nimodipine, and especially on a single large excitatory amino acids and blockade of the N-methyl-D- trial; 150 the evidence for nicardipine and AT877 was aspartate-glutamate receptor. It is also a non-competitive inconclusive (Fig. 14.15). antagonist of voltage-dependent calcium channels and The intermediate factors through which nimodipine dilates cerebral arteries. exerts its beneficial effect after aneurysmal SAH remain Two randomized controlled trials have studied intra- uncertain. Calcium entry-blocking drugs were tested venous magnesium sulphate in addition to nimodi- for this indication because they inhibit the contractile pine. The smallest included only 60 patients and was properties of smooth muscle cells, particularly in cere- inevitably inconclusive. 156 The larger trial included 283 bral arteries. Paradoxically, but not inconsistent with patients but was still intended as a preliminary (phase II) their effect on delayed cerebral ischaemia, several studies study with delayed cerebral ischaemia and not overall of nimodipine found there was no difference between clinical outcome as the primary measure of effective- treated patients and controls in the frequency of arterial ness. 157 Magnesium sulphate reduced delayed cerebral narrowing on a repeat angiogram. 150–152 Therefore, at ischaemia by 34% (hazard ratio 0.66; 95% CI 0.38–1.14). least with nimodipine, the protective effect on neurones After 3 months, the risk reduction for poor outcome seems to be more important than any reduction in vaso- was 23% (risk ratio 0.77; 95% CI 0.54–1.09). At that spasm (section 14.5.1). time, 18 patients in the treatment group and six in the The practical implication is that the regimen in the placebo group had an excellent outcome (risk ratio 3.4; dominant nimodipine trial (60 mg orally every 4 h, for 95% CI 1.3–8.9). A phase III trial has subsequently been 3 weeks) is currently regarded as the standard treat- launched with poor outcome as the primary measure of ment in patients with aneurysmal SAH. But, given the outcome (http://www.controlled-trials.com/mrct/trial/ uncertain effect on case fatality and the possibility that MAGNESIUM/2061/132459.html). Just one trial has the results of the meta-analysis might be affected by compared intravenous nimodipine with magnesium unpublished negative trials, the benefits of nimodipine sulphate, and found no difference in delayed cerebral cannot be regarded as being beyond all reasonable ischaemia, but the trial was too small (104 patients) to doubt. draw sound conclusions. 158 Oral nimodipine probably reduces the risk of a poor 14.5.4 Aspirin and other antithrombotic agents outcome after subarachnoid haemorrhage by about Several studies have found that platelets are activated one-third. from day 3 after subarachnoid haemorrhage, mostly .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 740 740 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage Study Treatment Control RR (fixed) Weight RR (fixed) or sub-category n/N n/N 95% CI (%) 95% CI 01 nimodipine, intravenously only Subtotal (95% CI) 0 0 Not estimable Total events: 0 (treatment), 0 (control) Test for heterogeneity: not applicable Test for overall effect: not applicable 02 nimodipine, intravenously followed by orally Ohman 1991 17/104 23/109 6.00 0.77 [0.44, 1.36] Han 1993 17/142 23/180 5.42 0.94 [0.52, 1.69] Subtotal (95%Cl) 246 246 11.41 0.85 [0.57, 1.28] Total events: 34 (treatment), 46 (control) 2 2 Test for heterogeneity: χ = 0.21, df = 1 (P = 0.65), I = 0% Test for overall effect: Z = 0.77 (P = 0.44) 03 nimodipine, orally only Neil-Dwyer 1987 9/38 17/37 4.60 0.52 [0.26, 1.01] Petruk 1988 44/72 54/82 13.48 0.93 [0.73, 1.18] Pickard 1989 55/278 91/276 24.39 0.60 [0.45, 0.80] Subtotal (95% Cl) 388 395 42.47 0.70 [0.58, 0.84] Total events: 108 (treatment), 162 (control) 2 2 Test for heterogeneity: χ = 7.26, df = 2 (P = 0.03), I = 72.5% Test for overall effect: Z = 3.77 (P = 0.0002) 04 nicardipine, intravenously Haley 1993 118/438 125/448 33.00 0.97 [0.78, 1.20] Subtotal (95% Cl) 438 448 33.00 0.97 [0.78, 1.20] Total events: 118 (treatment), 125 (control) Test for heterogeneity: not applicable Test for overall effect: Z = 0.32 (P = 0.75) 05 nicardipine, orally Subtotal (95% Cl) 0 0 Not estimable Total events: 0 (treatment), 0 (control) Test for heterogeneity: not applicable Test for overall effect: not applicable 06 AT877 Shibuya 1992 33/131 41/136 10.74 0.84 [0.57, 1.23] Subtotal (95% Cl) 131 136 10.74 0.84 [0.57, 1.23] Total events: 33 (treatment), 41 (control) Test for heterogeneity: not applicable Test for overall effect: Z = 0.90 (P = 0.37) 07 magnesium Veyna 2002 7/20 8/16 2.37 0.70 [0.32, 1.52] Subtotal (95% Cl) 20 16 2.37 0.70 [0.32, 1.52] Total events: 7 (treatment), 8 (control) Test for heterogeneity: not applicable Test for overall effect: Z = 0.90 (P = 0.37) Total (95% CI) 1223 100.00 0.82 [0.72, 0.93] Total events: 300 (treatment), 382 (control) 2 2 Test for heterogeneity: χ = 9.97, df = 7 (P = 0.19), I = 29.8% Test for overall effect: Z = 3.15 (P = 0.002) 0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control Fig. 14.15 Forest plot of the Cochrane review on the shown separately for each drug and route of administration. effectiveness of calcium antagonists on death and dependency, RR, relative risk; CI, confidence interval. 3–6 months after subarachnoid haemorrhage. 149 The results are .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 741 14.5 Prevention of delayed cerebral ischaemia 741 inferred from increased levels of thromboxane B2, and vascular effects. It has often been claimed that the stable metabolite of thromboxane A2, which pro- these ‘pleiotropic effects’ contribute to the reduction in motes platelet aggregation and vasoconstriction (section vascular events such as myocardial infarction beyond 16.5.3). 159,160 The question then is whether interventions that expected from LDL-cholesterol reduction alone, al- aimed at counteracting platelet activation are therapeu- though this has not been confirmed by a meta-regression tically useful. A retrospective analysis of 242 patients analysis of the randomized trials. 167 In patients with who had survived the first 4 days after SAH showed SAH, two randomized trials have been performed so far. that patients who had used salicylates before their hae- One included only 39 patients and found that 80 mg morrhage had a significantly decreased risk of delayed simvastatin given within 48 h of the onset reduced cerebral ischaemia, with or without permanent deficits ‘vasospasm’ (undefined), 168 and the other enrolled 80 (relative risk 0.40; 95% CI 0.18–0.93). 161 The first ran- patients and found that 40 mg pravastatin given within domized clinical trial was reported as early as 1982 but it 72 h reduced angiographic vasospasm and impair- failed to show any benefit from aspirin; 162 however, the ment of autoregulatory responses as well as vasospasm- number of patients was small, 53 unoperated patients related ischaemic complications. 169 On the other hand, were also included, and all were treated with tranexamic an observational study found that previous use of statins acid which increases the risk of ischaemia (section increased the risk of angiographic vasospasm though not 14.4.4). A further study of aspirin after early operation in that of associated ischaemic complications. 170 In conclu- 50 patients confirmed that this treatment was feasible sion, the evidence for a beneficial effect of statins after and probably safe. 163 The efficacy, however, could not be SAH is still very meagre. confirmed in a randomized trial of aspirin given after occlusion of the aneurysm that aimed to include 200 14.5.6 Other drugs patients; this was prematurely stopped after the second interim analysis because by then the chances of a posit- Tirilazad mesylate, a 21-aminosteroid free radical scav- ive effect were negligible. At the final analysis, aspirin enger, has not consistently improved outcome in four did not reduce the risk of delayed ischaemia (hazard ratio randomized controlled trials, with a total of more than (HR) 1.83; 95% CI 0.8–3.9) nor that of poor outcome (HR 3500 patients. 171–174 Although, in one single subgroup (the 0.79; 95% CI 0.4–1.6). 95 one with the highest dosage) a beneficial effect on over- Three other antiplatelet drugs have been tested in all outcome was seen, 171 this could not be reproduced in patients with SAH: dipyridamole; 164 the thromboxane the corresponding subgroup from a parallel trial, 172 nor A2 synthetase inhibitor, cataclot; 165 and the experi- in two subsequent trials. 173,174 In these last two trials, mental antiplatelet agent OKY-46. 166 Ischaemic neuro- only women were included and they were treated with logical deficits were reduced in one 166 but not in the other even higher dosages than in the preceding trials, because trial that assessed this outcome event. 165 In a systematic in the first two trials women had appeared to respond overview of these three trials and the three published less than men. A paper reporting that patients on the aspirin trials, the risk of delayed cerebral ischaemia study drug in a small subgroup from New Zealand had (reported in only four of the five trials) was decreased, less fatigue and better neuropsychological performance but not to a statistically significant level (relative risk than controls suggests desperation rather than con- 0.81; 95% CI 0.6–1.1). For the three aspirin studies viction among the investigators and the sponsoring alone, the relative risk for delayed cerebral ischaemia company. 175 was 1.34 (95% CI 0.8–2.3). Poor clinical outcome was A single trial with another hydroxyl radical scavenger, not significantly different between patients treated with N′-propylenedinicotinamide (nicaraven), in 162 patients, antiplatelet drugs and controls (relative risk 0.86; 95% CI showed a decrease in delayed cerebral ischaemia but 0.6–1.3) (Dorhout Mees et al., unpublished data). not in poor clinical outcome at 3 months after SAH. 176 A low-molecular-weight heparinoid, enoxoparin has Curiously enough, the reverse was found in a trial of 286 been tested in a trial of 170 patients but it did not improve patients with ebselen, a seleno-organic compound with outcome and was associated with intracranial haemor- antioxidant activity through a glutathione peroxidase- rhage in 4 of 85 patients in the experimental group. 56 like action. 177 A formal overview of all the evidence for free radical scavengers is not yet available, but the case for these drugs seems weak. 14.5.5 Statins Nizofenone, an anionic channel blocker believed to HMG-CoA reductase inhibitors, or statins, are primarily inhibit glutamate release, was studied in a randomized used to lower LDL-cholesterol levels but they also have trial of 100 patients, of whom only 90 were included anti-inflammatory, immunomodulatory, antithrombotic in the analysis; 178 angiographic vasospasm and poor .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 742 742 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage outcome were not affected, and an effect on poor out- 14.5.8 Cisternal drainage and intracisternal come was found only in an on-treatment analysis in the fibrinolysis subgroup of patients with vasospasm. The endothelin receptor antagonist TAK-044 was On the assumption that vasospasm increases the risk A/B tested in a multicentre phase II trial in 420 patients; 179 of delayed cerebral ischaemia and that extravasated there was a non-significant risk reduction of 0.8 in blood induces vasospasm, removal of the subarachnoid delayed ischaemic deficits (95% CI 0.61–1.06). blood by drainage or fibrinolysis has been suggested and then studied in several trials. For example, in a non- randomized comparison, patients treated with lumbar 14.5.7 Increasing the plasma volume drainage of CSF had cerebral infarction less often and The usefulness of circulatory volume expansion to pre- more often returned home than patients with no lumbar vent delayed cerebral ischaemia after SAH has been drainage. 185 Because no randomized trials have been assessed in a recent Cochrane review. 180 Only one truly performed yet, this strategy cannot be recommended for randomized and one quasi-randomized trial with com- routine clinical practice. An even more aggressive way to parable baseline characteristics for both groups could remove subarachnoid blood is intracisternal fibrinolysis. be included in the analyses. Volume expansion did not A meta-analysis of this treatment strategy included improve outcome (relative risk [RR] 1.0; 95% CI 0.5–2.2), nine trials of which only one was randomized. 186 Pooled nor delayed cerebral ischaemia (RR 1.1; 95% CI 0.5–2.2), results demonstrated beneficial effects of treatment but but tended to increase the complications (RR 1.8; 95% the conclusions for practice are obviously very limited by CI 0.9–3.7). In another quasi-randomized trial, not the predominance of non-randomized studies. An open, included in the analyses, outcome was assessed only on randomized, controlled trial not yet included in the the day of operation (7–10 days after SAH). In the period meta-analysis tested fibrinolysis in 110 patients treated before operation, treatment resulted in a reduction of with endovascular coiling. 187 Urokinase was adminis- secondary ischaemia (RR 0.33; 95% CI 0.11–0.99) and tered into the cisterna magna through a microcatheter case fatality (RR 0.20; 95% CI 0.07–1.2). In conclusion, inserted via a lumbar puncture. There was a statistically volume expansion has been studied properly in only significant improvement in the primary outcome of two trials of patients with aneurysmal SAH, with very clinical vasospasm, defined as clinical deterioration small numbers, and so far there is no sound evidence to combined with evidence of vasospasm on angiography, support this therapy. but case fatality was not reduced even though patients Because of its mineralocorticoid activity (reabsorption in the treated group more often had a good clinical of sodium in the distal tubules of the kidney) fludrocorti- outcome. Larger studies with overall clinical outcome sone might, in theory, prevent a negative sodium bal- as the primary measurement of outcome are needed ance, hypovolaemia and so ischaemic complications. 181 before this treatment can be implemented in clinical A randomized study in 91 patients with SAH showed practice. that although fludrocortisone acetate did reduce natri- uresis in the first 6 days after the haemorrhage, there was no definite effect on plasma volume depletion or on ischaemic complications, although any beneficial effects may have been masked because patients in the control 14.6 Management of rebleeding group were often treated with plasma expanders after they had developed clinical signs of ischaemia. 182 These results were confirmed by a smaller trial in 30 patients. 183 14.6.1 Diagnosis Finally, hydrocortisone was also shown in a small trial of 28 patients to prevent hyponatraemia and a drop Loss of consciousness was the cardinal feature of rebleed- in central venous pressure. 184 The evidence from these ing in a prospective series of 39 patients with CT-proven studies is insufficiently conclusive to warrant routine rebleeding: of the 36 patients who were awake at the fludrocortisone for all patients with SAH. time, 35 lost consciousness, with preceding headache in one-third; in the remaining patient a sudden increase Although it is reasonable to prevent a decrease in 188 in headache was the only symptom. Serial CT scan- plasma volume, there is no good evidence to support ning uncovered rebleeding in two of the three remaining prevention of cerebral ischaemia after SAH through patients, in whom the level of consciousness was already increasing plasma volume by infusion of albumin or minimal. No ‘silent rebleeding’ was found in awake colloids, or by the administration of fludrocortisone. patients. In patients with a first episode of subarachnoid .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 743 14.6 Management of rebleeding 743 Table 14.5 Specific causes of sudden deterioration in patients usually become clear within a matter of hours whether with aneurysmal subarachnoid haemorrhage. 383 the patient will indeed survive the episode or whether dysfunction of the brainstem will persist. There are no Rebleeding (two-thirds) (section 14.6.1) grounds to fear that resuscitation will only result in pro- Epileptic seizure (section 14.3.3) longation of a vegetative state and, in patients who do Delayed cerebral ischaemia, with atypical, sudden onset progress to a state of brain death, the resuscitation pro- (section 14.7) cedure at least allows organ donation to be considered, Ventricular fibrillation (section 14.9.3) with some benefits to others. Sudden apnoea or cardiac arrest in a patient with haemorrhage, loss of consciousness is less common, subarachnoid haemorrhage usually signifies occurring in slightly less than half. 55 That unconscious- rebleeding. With full resuscitative measures cardiac ness occurs more often with rebleeding than with a function usually recovers within minutes and first rupture confirms that rebleeding has a more severe spontaneous respiration will rapidly return in impact on the brain, and so a higher risk of a poor approximately 50% of the cases, followed within 4 outcome (80%). In fatal episodes of rebleeding most hours by return of consciousness and of brainstem patients immediately lose their brainstem reflexes at reflexes if these had been lost at the same time. the same time as consciousness, but without initial If spontaneous respiration does not return, then apnoea. organ donation should be considered. However, rebleeding is not the only cause of a sudden deterioration of consciousness in a patient with a rup- 14.6.3 Emergency occlusion of the aneurysm tured aneurysm; other complications underlie about one-third of such episodes (Table 14.5). Therefore, once A large haematoma that causes brain shift without gross the respiratory and circulatory state of the patient has intraventricular haemorrhage is infrequent after rebleed- sufficiently stabilized, it is mandatory to confirm the ing, occurring in around 10% of patients. 189 In these rare diagnosis of rebleeding by repeat brain CT. cases, emergency evacuation of the haematoma after aneurysm occlusion may be indicated, as after first rup- ture (section 14.2.2). A more common reason for urgent 14.6.2 To resuscitate or not? aneurysm occlusion after rebleeding is the concern that, Whether patients with rebleeding should be resuscitated among the survivors, 50–75% will have further episodes and ventilated if respiratory arrest occurs is not aca- of rebleeding if the aneurysm is left untreated. 9,188 This demic; in the series of 39 patients with CT-confirmed implies that emergency clipping or coiling of the aneurysm rebleeding, 14 had initial respiratory abnormalities that should be seriously considered in patients who regain called for assisted ventilation. Spontaneous respiration consciousness after rebleeding. Of course, the risk of the returned within 1 h in eight of these 14 patients, and in operation is increased after rebleeding but the risks of a three more between 1 and 24 h. 188 In another study of wait-and-see policy at that stage seem even more intim- episodes of respiratory arrest but in which first bleeds idating. The management of rebleeding is summarized were also included, whether the patient would or would in Table 14.6. not regain spontaneous respiration could not be pre- dicted from the anatomical site of haemorrhage on CT, the initial presence or absence of brainstem reflexes, or the type of respiratory disorder. 12 Many patients with Table 14.6 Management of rebleeding. initial apnoea who were successfully resuscitated later died from subsequent complications, but survival with- If cardiac or respiratory arrest, resuscitate and ventilate; out brain damage is clearly possible – even after respir- cardiac resuscitation is usually successful and within atory arrest. hours either spontaneous respiration will return or all Rebleeding can also cause cardiac arrest, but again in brainstem functions will be lost this situation resuscitation is worthwhile. Cardiac func- Repeat CT brain scan to confirm diagnosis tion usually recovers within minutes and some patients Consider emergency clipping or coiling of aneurysm can have a good overall outcome. Out of 11 patients after recovery, because many patients will have further episodes of rebleeding, with a high case fatality with 14 episodes of well-documented cardiac arrest, Large intracerebral haematomas can be evacuated at the six survived and three of them became independent same time for activities of daily living. 13 After resuscitation, it will .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 744 744 Chapter 14 Specific treatment of aneurysmal subarachnoid haemorrhage Table 14.7 Causes of gradual deterioration after subarachnoid extent and location of the ischaemia. A recent study haemorrhage. distinguished two patterns of infarction: single cortical infarcts, typically near the ruptured aneurysm, and Oedema surrounding intracerebral haematoma multiple widespread lesions including subcortical loca- Delayed cerebral ischaemia (section 14.7.1) tions and often unrelated to the site of aneurysm Hydrocephalus (section 14.8.1) 143 rupture. Unsuspected rebleeding (section 14.6.1) MRI is more sensitive in detecting early changes in the Systemic complication (section 14.9) water content of the brain, especially with diffusion- Hyponatraemia Disturbance of heart rhythm weighted imaging, 191 but the acquisition time needed to Hypoxia from pulmonary oedema or chest infection obtain good images is often too long for ill and restless Hypotension patients. Single-photon emission CT (SPECT) scanning Infection may be helpful but hypoperfusion may also result from oedema around a resolving haematoma or, in the basal parts of the brain, from hydrocephalus. 192,193 Transcranial Doppler sonography (TCD) may suggest 14.7 Management of delayed cerebral impending cerebral ischaemia by the increased blood ischaemia flow velocity from arterial narrowing, but there is con- siderable overlap with patients who do not develop ischaemia. A systematic review of studies in which TCD findings were compared with vascular imaging found 14.7.1 Diagnosis that only for the middle cerebral artery was there good The diagnosis of delayed cerebral ischaemia after SAH is specificity (99%; 95% CI 98–100%) and moderate sensit- far less straightforward than diagnosing ischaemic stroke, ivity (67%; 95% CI 48–87%). 194 For other arteries, TCD not only in daily practice, but also in research. To com- was neither accurate nor useful. However, most of these pound the problem, in many papers delayed cerebral data are of low methodological quality, bias cannot be ischaemia is poorly defined or not even defined at all. 190 ruled out and the data reporting was often uncritical. This difficulty in recognising reliably delayed cerebral Furthermore, demonstration of arterial narrowing does ischaemia is disturbing because it is only one of several not prove in itself that clinical deterioration has been other causes of gradual deterioration in patients after caused by ischaemia. Factors such as intracranial pres- SAH (Table 14.7). sure, arterial blood pressure and respiration may all The clinical manifestations of delayed cerebral influence the form of the velocity wave. ischaemia evolve gradually, over several hours, and usu- Multi-slice CT angiography provides results that are in ally between day 4 and 12 after the haemorrhage. In good agreement with those of catheter angiography, 195,196 25% of the patients ischaemia causes hemispheric focal but the question still remains how useful is it to obtain deficits, in another 25% a decrease in the level of con- information about arterial narrowing, given the imper- sciousness, and in the remaining 50% these two features fect correlation with the occurrence of brain ischaemia. develop at the same time. 128 The diagnosis should be In a study where vasospasm was diagnosed by TCD and based not only on this clinical picture, but also on the catheter angiography, its presence was associated with results of laboratory investigations and a repeat brain CT delayed cerebral ischaemia in only two-thirds of patients to exclude other intracranial causes of subacute deter- (positive predictive value 0.67) and its absence with no ioration, such as oedema surrounding an intracerebral delayed cerebral ischaemia in a similar proportion (neg- haematoma, dilatation of the ventricular system or ative predictive value 0.72). 138 unsuspected rebleeding. Positive evidence of ischaemia Because delayed cerebral ischaemia is difficult to dia- is shown by CT in about 80% of patients after exclu- gnose from the outset, many physicians rely on TCD sion of these other causes, at least after a few days from or angiographic studies to rule in or out the presence of the onset of the clinical signs of delayed cerebral vasospasm. However, since the negative and positive ischaemia. 128 The disadvantage of CT scanning is that it predictive values of vasospasm for cerebral ischaemia are confirms cerebral ischaemia only some time after the moderate at best, knowing whether there is or is not clinical deterioration, which is too late for interventions vasospasm is not very helpful. In fact, ordering a tran- that may be applied early after the onset of the ischaemia. scranial Doppler (or even worse catheter angiography) However, currently CT scanning is the only method to in a patient suspected of delayed cerebral ischaemia confirm delayed cerebral ischaemia. One advantage of is comparable to ordering a duplex of the carotid artery CT scanning is that it provides information about the to confirm by the presence or absence of carotid artery .. ..
9781405127660_4_014.qxd 10/13/07 2:15 PM Page 745 14.7 Management of delayed cerebral ischaemia 745 stenosis a transient ischaemic attack as the cause of an Table 14.8 Management of cerebral ischaemia. episode of unclear neurological symptoms. Immediately give 500 mL of colloid solution i.v. Consider subclavian vein catheter and maintain central 14.7.2 Induced hypertension and volume venous pressure between 8 and 12 mmHg expansion Consider increasing mean arterial pressure by 20–40 mmHg above baseline values Since the 1960s, induced hypertension has been used to Correct any hypoxaemia combat ischaemic deficits in patients with SAH. Particu- Maintain fluid intake with at least 3 L of normal saline larly remarkable was the report of a patient in whom (0.9%) per 24 h postoperative aphasia and right hemiparesis fluctuated Correct hyponatraemia if severe (Table 14.10) with the level of the blood pressure. 197 Later, induced hypertension was often combined with volume expan- sion, but again only in uncontrolled case series, of which just one had an acceptable definition of outcome values. These procedures should be carried out only in an events. 198 Another case series argued that hypertensive intensive care unit with facilities for specialized care and and hypervolaemic therapy was unlikely to be successful close monitoring. in patients with a Glasgow Coma Score of 11 or less, as well as with hydrocephalus. 199 However, on the basis of 14.7.3 Transluminal angioplasty and vasodilating these and other case reports and series many physicians drugs have used induced hypertension and hypervolaemia and seen their patients improve, but randomized controlled Only a few centres have reported on endovascular treat- trials are sadly missing. ment of ‘symptomatic vasospasm’ after SAH, claiming If raising the blood pressure and increasing plasma sustained improvement in more than half the cases, but volume can indeed reverse ischaemic deficits – which these series were uncontrolled and there must be pub- remains to be proven beyond doubt – the most plausible lication bias. 205 Some of these studies reported results explanation is a defect of cerebral autoregulation that only for arteries, and not – more relevantly – for patients. makes the perfusion of the injured brain passively depend- Vessel rupture is precipitated by this procedure in about ent on the systemic blood pressure. To add haemodilu- 1%, even after the aneurysm has been occluded, and tion to hypertensive and hypervolaemic treatment (the other complications such as hyperperfusion injury so-called ‘triple-H’ regimen) is not only of uncertain in 4%. 205 In another uncontrolled series of patients benefit, but even controversial. 200 treated with transluminal angioplasty, papaverine injec- The risks of deliberately increasing the arterial pres- tion or both, overall clinical outcome was poor despite sure and plasma volume include rebleeding of untreated successful arterial dilatation. Half the patients died or aneurysms additional to the ruptured aneurysm, although remained disabled, and half the survivors had perman- this procedure did not in fact cause haemorrhages in ent deficits from cerebral infarction. 206 In view of the a series of 40 patients with 73 unsecured additional risks, the high costs and the lack of controlled trials, aneurysms. 201 Other complications are increased cere- transluminal angioplasty should still be regarded as a bral oedema, haemorrhagic transformation in areas of strictly experimental procedure. 207 infarction, 202 reversible leucencephalopathy, 203 myocar- The same caution applies to uncontrolled reports of dial infarction and congestive heart failure. Certainly the improvement of ischaemic deficits after intra-arterial circulatory system should be closely monitored, though infusion of drugs through super-selective catheteriza- arterial lines and pulmonary artery catheters carry their tion. Papaverine has gained undeserved popularity, 208 the own risks: infection, pneumothorax, haemothorax, ven- more so because not all impressions are positive. 209,210 tricular arrhythmia and pulmonary infarction. 48–50 Intra-arterial milrinone, verapamil or nicardipine dilate A currently recommended regimen (Table 14.8), vessels but whether they improve clinical outcome is although not supported by rigorously controlled trials, is very uncertain. 211–213 to start with plasma volume expansion with Hetastarch Calcitonin gene-related peptide (CGRP) is a potent or another colloid solution (5% albumin does not vasodilator in the carotid vascular bed, but a random- increase plasma volume). 204 The aim is to raise the ized, multicentre, single-blind clinical trial in 62 patients central venous filling pressure to approximately 8–12 with ischaemic complications after SAH failed to show mmHg. If there is no clinical improvement, one might any benefit in terms of overall clinical outcome: the rel- consider raising the blood pressure with dopamine ative risk of a poor outcome in CGRP-treated patients or dobutamine by 20–40 mmHg above pre-treatment was 0.88 (95% CI 0.60–1.28). 214 .. ..
Search
Read the Text Version
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
- 13
- 14
- 15
- 16
- 17
- 18
- 19
- 20
- 21
- 22
- 23
- 24
- 25
- 26
- 27
- 28
- 29
- 30
- 31
- 32
- 33
- 34
- 35
- 36
- 37
- 38
- 39
- 40
- 41
- 42
- 43
- 44
- 45
- 46
- 47
- 48
- 49
- 50
- 51
- 52
- 53
- 54
- 55
- 56
- 57
- 58
- 59
- 60
- 61
- 62
- 63
- 64
- 65
- 66
- 67
- 68
- 69
- 70
- 71
- 72
- 73
- 74
- 75
- 76
- 77
- 78
- 79
- 80
- 81
- 82
- 83
- 84
- 85
- 86
- 87
- 88
- 89
- 90
- 91
- 92
- 93
- 94
- 95
- 96
- 97
- 98
- 99
- 100
- 101
- 102
- 103
- 104
- 105
- 106
- 107
- 108
- 109
- 110
- 111
- 112
- 113
- 114
- 115
- 116
- 117
- 118
- 119
- 120
- 121
- 122
- 123
- 124
- 125
- 126
- 127
- 128
- 129
- 130
- 131
- 132
- 133
- 134
- 135
- 136
- 137
- 138
- 139
- 140
- 141
- 142
- 143
- 144
- 145
- 146
- 147
- 148
- 149
- 150
- 151
- 152
- 153
- 154
- 155
- 156
- 157
- 158
- 159
- 160
- 161
- 162
- 163
- 164
- 165
- 166
- 167
- 168
- 169
- 170
- 171
- 172
- 173
- 174
- 175
- 176
- 177
- 178
- 179
- 180
- 181
- 182
- 183
- 184
- 185
- 186
- 187
- 188
- 189
- 190
- 191
- 192
- 193
- 194
- 195
- 196
- 197
- 198
- 199
- 200
- 201
- 202
- 203
- 204
- 205
- 206
- 207
- 208
- 209
- 210
- 211
- 212
- 213
- 214
- 215
- 216
- 217
- 218
- 219
- 220
- 221
- 222
- 223
- 224
- 225
- 226
- 227
- 228
- 229
- 230
- 231
- 232
- 233
- 234
- 235
- 236
- 237
- 238
- 239
- 240
- 241
- 242
- 243
- 244
- 245
- 246
- 247
- 248
- 249
- 250
- 251
- 252
- 253
- 254
- 255
- 256
- 257
- 258
- 259
- 260
- 261
- 262
- 263
- 264
- 265
- 266
- 267
- 268
- 269
- 270
- 271
- 272
- 273
- 274
- 275
- 276
- 277
- 278
- 279
- 280
- 281
- 282
- 283
- 284
- 285
- 286
- 287
- 288
- 289
- 290
- 291
- 292
- 293
- 294
- 295
- 296
- 297
- 298
- 299
- 300
- 301
- 302
- 303
- 304
- 305
- 306
- 307
- 308
- 309
- 310
- 311
- 312
- 313
- 314
- 315
- 316
- 317
- 318
- 319
- 320
- 321
- 322
- 323
- 324
- 325
- 326
- 327
- 328
- 329
- 330
- 331
- 332
- 333
- 334
- 335
- 336
- 337
- 338
- 339
- 340
- 341
- 342
- 343
- 344
- 345
- 346
- 347
- 348
- 349
- 350
- 351
- 352
- 353
- 354
- 355
- 356
- 357
- 358
- 359
- 360
- 361
- 362
- 363
- 364
- 365
- 366
- 367
- 368
- 369
- 370
- 371
- 372
- 373
- 374
- 375
- 376
- 377
- 378
- 379
- 380
- 381
- 382
- 383
- 384
- 385
- 386
- 387
- 388
- 389
- 390
- 391
- 392
- 393
- 394
- 395
- 396
- 397
- 398
- 399
- 400
- 401
- 402
- 403
- 404
- 405
- 406
- 407
- 408
- 409
- 410
- 411
- 412
- 413
- 414
- 415
- 416
- 417
- 418
- 419
- 420
- 421
- 422
- 423
- 424
- 425
- 426
- 427
- 428
- 429
- 430
- 431
- 432
- 433
- 434
- 435
- 436
- 437
- 438
- 439
- 440
- 441
- 442
- 443
- 444
- 445
- 446
- 447
- 448
- 449
- 450
- 451
- 452
- 453
- 454
- 455
- 456
- 457
- 458
- 459
- 460
- 461
- 462
- 463
- 464
- 465
- 466
- 467
- 468
- 469
- 470
- 471
- 472
- 473
- 474
- 475
- 476
- 477
- 478
- 479
- 480
- 481
- 482
- 483
- 484
- 485
- 486
- 487
- 488
- 489
- 490
- 491
- 492
- 493
- 494
- 495
- 496
- 497
- 498
- 499
- 500
- 501
- 502
- 503
- 504
- 505
- 506
- 507
- 508
- 509
- 510
- 511
- 512
- 513
- 514
- 515
- 516
- 517
- 518
- 519
- 520
- 521
- 522
- 523
- 524
- 525
- 526
- 527
- 528
- 529
- 530
- 531
- 532
- 533
- 534
- 535
- 536
- 537
- 538
- 539
- 540
- 541
- 542
- 543
- 544
- 545
- 546
- 547
- 548
- 549
- 550
- 551
- 552
- 553
- 554
- 555
- 556
- 557
- 558
- 559
- 560
- 561
- 562
- 563
- 564
- 565
- 566
- 567
- 568
- 569
- 570
- 571
- 572
- 573
- 574
- 575
- 576
- 577
- 578
- 579
- 580
- 581
- 582
- 583
- 584
- 585
- 586
- 587
- 588
- 589
- 590
- 591
- 592
- 593
- 594
- 595
- 596
- 597
- 598
- 599
- 600
- 601
- 602
- 603
- 604
- 605
- 606
- 607
- 608
- 609
- 610
- 611
- 612
- 613
- 614
- 615
- 616
- 617
- 618
- 619
- 620
- 621
- 622
- 623
- 624
- 625
- 626
- 627
- 628
- 629
- 630
- 631
- 632
- 633
- 634
- 635
- 636
- 637
- 638
- 639
- 640
- 641
- 642
- 643
- 644
- 645
- 646
- 647
- 648
- 649
- 650
- 651
- 652
- 653
- 654
- 655
- 656
- 657
- 658
- 659
- 660
- 661
- 662
- 663
- 664
- 665
- 666
- 667
- 668
- 669
- 670
- 671
- 672
- 673
- 674
- 675
- 676
- 677
- 678
- 679
- 680
- 681
- 682
- 683
- 684
- 685
- 686
- 687
- 688
- 689
- 690
- 691
- 692
- 693
- 694
- 695
- 696
- 697
- 698
- 699
- 700
- 701
- 702
- 703
- 704
- 705
- 706
- 707
- 708
- 709
- 710
- 711
- 712
- 713
- 714
- 715
- 716
- 717
- 718
- 719
- 720
- 721
- 722
- 723
- 724
- 725
- 726
- 727
- 728
- 729
- 730
- 731
- 732
- 733
- 734
- 735
- 736
- 737
- 738
- 739
- 740
- 741
- 742
- 743
- 744
- 745
- 746
- 747
- 748
- 749
- 750
- 751
- 752
- 753
- 754
- 755
- 756
- 757
- 758
- 759
- 760
- 761
- 762
- 763
- 764
- 765
- 766
- 767
- 768
- 769
- 770
- 771
- 772
- 773
- 774
- 775
- 776
- 777
- 778
- 779
- 780
- 781
- 782
- 783
- 784
- 785
- 786
- 787
- 788
- 789
- 790
- 791
- 792
- 793
- 794
- 795
- 796
- 797
- 798
- 799
- 800
- 801
- 802
- 803
- 804
- 805
- 806
- 807
- 808
- 809
- 810
- 811
- 812
- 813
- 814
- 815
- 816
- 817
- 818
- 819
- 820
- 821
- 822
- 823
- 824
- 825
- 826
- 827
- 828
- 829
- 830
- 831
- 832
- 833
- 834
- 835
- 836
- 837
- 838
- 839
- 840
- 841
- 842
- 843
- 844
- 845
- 846
- 847
- 848
- 849
- 850
- 851
- 852
- 853
- 854
- 855
- 856
- 857
- 858
- 859
- 860
- 861
- 862
- 863
- 864
- 865
- 866
- 867
- 868
- 869
- 870
- 871
- 872
- 873
- 874
- 875
- 876
- 877
- 878
- 879
- 880
- 881
- 882
- 883
- 884
- 885
- 886
- 887
- 888
- 889
- 890
- 891
- 892
- 893
- 894
- 895
- 896
- 897
- 898
- 899
- 900
- 901
- 902
- 903
- 904
- 905
- 906
- 907
- 908
- 909
- 910
- 911
- 912
- 913
- 914
- 915
- 916
- 917
- 918
- 919
- 920
- 921
- 922
- 923
- 924
- 925
- 926
- 927
- 928
- 929
- 930
- 931
- 932
- 933
- 934
- 935
- 936
- 937
- 938
- 939
- 940
- 941
- 942
- 943
- 944
- 945
- 946
- 947
- 948
- 949
- 950
- 951
- 952
- 953
- 954
- 955
- 956
- 957
- 958
- 959
- 960
- 961
- 962
- 963
- 964
- 965
- 966
- 967
- 968
- 969
- 970
- 971
- 972
- 973
- 974
- 975
- 976
- 977
- 978
- 979
- 980
- 981
- 982
- 983
- 984
- 985
- 986
- 987
- 988
- 989
- 990
- 991
- 992
- 993
- 994
- 995
- 996
- 997
- 998
- 1 - 50
- 51 - 100
- 101 - 150
- 151 - 200
- 201 - 250
- 251 - 300
- 301 - 350
- 351 - 400
- 401 - 450
- 451 - 500
- 501 - 550
- 551 - 600
- 601 - 650
- 651 - 700
- 701 - 750
- 751 - 800
- 801 - 850
- 851 - 900
- 901 - 950
- 951 - 998
Pages: