You are on page 1of 7

Effect of Etiology and Topography of Lesion on Body Temperature at Stroke Onset

Emre Kumral,
MD,

Sultan Tarlaci,

MD,

and Ahmet Acarer,

MD

Background and purpose: Hyperthermia is a well-known factor for neurologic deterioration, morbidity, and mortality in the early phase of stroke. However, the timing, localization of lesion, origin of stroke, which may inuence body temperature, have not been clearly established. Methods: The purpose of this study was to determine the relationship between body temperature and origin, lesion topography, and prognosis at 3 months after onset of stroke. Axillary temperature was taken every hour for 72 hours in 473 patients with supra- or infratentorial cerebral vascular lesion. The time at which hyperthermia (38C) appeared was evaluated by logistic regression analyses regarding to stroke origin and lesion localization. The correlation between body temperature and stroke outcome was quantied by Barthel index and American Heart Association Stroke Outcome Classication by recording in each 12- hour interval from stroke onset during 72 hours and after 3 months. Results: The body temperature was higher in patients with large-artery atherosclerosis (odds ratio [OR], 3.98; 95% condence interval [CI] 2.16-8.97; P .001) and hemorrhagic stroke (OR 2.05, 95% CI 1.07-8.68, P .001) than those with small-artery disease. In patients with posterior circulation infarct, the body temperature was higher than those with anterior circulation infarct (OR 3.71, 95% CI 2.07-6.67, P .001), whereas there was no difference between patients with infratentorial hemorrhage and those with supratentorial hemorrhage (OR 1.04, 95% CI 0.75-1.43, P .80). High body temperature at 24 hours of stroke onset (OR 2.17, 95% CI 2.09-7.57, P .001) and 48 hours (OR 1.27, 95% CI 1.06-4.84, P .02) was correlated with poor outcome and mortality. Conclusion: An association between hyperthermia within 72 hours of ictus and stroke subtypes was observed among patients with ischemic and hemorrhagic stroke. Hyperthermic patients with total anterior circulation infarct, posterior circulation infarct, and supratentorial hemorrhage were associated with a marked increase of 3-months mortality. Large-artery atherosclerosis, cardioembolism, and supra-infratentorial hemorrhage associated with hyperthermia may increase the severity of neurologic decits. Copyright 2001 by National Stroke Association

Recent studies have claried that changes in body temperature can inuence long-term prognosis of patients with acute stroke.1-10 Early laboratory ndings in
From the Department of Neurology, Ege University, Izmir, Turkey. Received January 26, 2001; accepted May 22, 2001. Address reprint requests to Prof. Emre Kumral, MD, Stroke Unit, Department of Neurology, Ege University Faculty of Medicine, Bornova, 35100, Izmir, Turkey. Copyright 2001 by National Stroke Association 1052-3057/01/1004-0004$35.00/0 doi:10.1053/jscd.2001.26867

animal models showed that small elevations of body temperature after induced cerebral infarction may cause increase in cerebral infarct volume.11-13 Similarly, in humans, 0.5 to 1.4C temperature increases were associated with a worse outcome in humans after acute stroke.14 A recent meta-analysis of 9 studies evaluating the effects of post-stroke hyperthermia on stroke outcome suggested that pyrexia after stroke onset is associated with a marked increase in morbidity and mortality.15 Regarding these results, hypothermia was entered in open-heart surgery and neurosurgery to combat the detrimental effects of cerebral hypoxic or ischemic events.16,17

150

Journal of Stroke and Cerebrovascular Diseases, Vol. 10, No. 4 (July-August), 2001: pp 150-156

EFFECT OF ORIGIN AND TOPOGRAPHY

151

We sought the relationship between body temperature at acute stroke and stroke subtypes, including topography of lesion. The effect of body temperature on stroke outcome regarding the cause and localization of lesion was also studied.

Materials and Methods


The study subgroup was recruited from 1097 stroke patients (913 with ischemic stroke, 184 with intracerebral hemorrhage) admitted consecutively between April 1997 and November 1998. We analyzed only 473 stroke patients without a source of infection and who were eligible for the study design. The inclusion criteria were the following: (1) admission less than 4 hours from onset of symptoms; (2) presence of parenchymal hemorrhage on the cranial computed tomography (CT) scan; (3) presence of neurologic decit at admission with or without sign of hypodensity or obscuration of brain tissue on the CT scan; (4) lack of pulmonary and urinary infections at the time of hospital admission, deep venous thrombosis, and cancer; and (5) lack of treatment with sympathomimetics, antidepressants, antiepileptics, and antibiotics. Patients without detailed medical records or without CT scan or magnetic resonance imaging (MRI) and with primary subarachnoid hemorrhage or subdural hemorrhage were excluded. On admission, axillary temperature was recorded every hour after admission during the rst 3 days. For the aim of this study, only the time at which hyperthermia appeared was recorded each hour from admission. Hyperthermia was dened as an axillary temperature above 38C. Blood analyses, chest radiograph, hemocultures, and urine cultures were performed to exclude a potential infection. Hyperthermia was considered to be of infectious origin if the patients had positive test results after stroke onset, and the patients were not included in the study. We treated patients who had body temperature 38C with antipyretics or with external cooling during the hospitalization. We performed 2 to 5 CT examinations in different stroke patients depending on their neurologic status. Noninvasive investigations that were performed in all cases included transcranial Doppler sonography; duplex sonography; and cardiac investigations consisting of a clinical examination, 12-lead electrocardiogram, as well as transthoracic/transesophageal echocardiography. The following vascular risk factors were recorded: age, sex, hypertension (on 2 or more blood pressure recordings higher than 160/90 mm Hg), diabetes mellitus (fasting serum glucose level higher than 6.5 mM/L), hypercholesterolemia (fasting blood cholesterol higher than 6.5 mM/L), cigarette smoking, ischemic heart disease, atrial brillation, and cardiac valve prosthesis.

Topography of ischemic stroke was classied into 4 main groups based on the denitions of the Trial of ORG 10172 in Acute Stroke Therapy (TOAST) criteria.18 They were as follows: (1) total anterior circulation infarct was presumed in patients with hemiplegia, hemianopia, and disturbance of higher cerebral function; (2) partial anterior circulation infarct was presumed in those with sensory motor decit plus hemianopia or higher cerebral dysfunction; (3) posterior circulation infarct was presumed in patients with ipsilateral cranial nerve palsy with contralateral motor/sensory decits or disorder of conjugate gaze palsy or cerebellar dysfunction or isolated hemianopia; and (4) lacunar infarct (or small-artery disease) was considered in patients having pure motor stroke, pure sensory stroke, homolateral ataxia and crural paresis, and dysarthia clumsy-hand syndrome. The cause of stroke was divided into ischemic and hemorrhagic stroke (or primary intracerebral hemorrhage). The origin of ischemic stroke was classied as follows: (1) largeartery atherosclerosis (LAA) was presumed in patients who had a stenosis of at least 30% of the lumen diameter in the appropiate large artery as shown on Duplex, transcranial Doppler, or magnetic resonance angiography in the absence of other origins; (2) small-artery disease (SAD) was presumed in patients with longstanding hypertension or diabetes mellitus, and an infarction with a diameter of 15 mm limited to the territory of deep perforators on CT scan or MRI in the absence of other etiologies; (3) cardiac embolic sources including nonvalvular atrial brillation, left ventricular dyskinetic segment, intracardiac thrombus or tumor, mitral stenosis, and other less common sources; (4) mixed origin was assumed in cases of coexistence of large-artery disease and cardiac embolic sources; and (5) undetermined and other causes, which were not compatible with largeartery atheroma or cardiac embolic sources or remained undetermined. The origin of hemorrhagic stroke was considered as hypertension, vascular malformations, anticoagulation, amiloid angiopathy, and others. Intracerebral hemorrhage topography was subdivided into 2 main subgroups: (1) supratentorial hemorrhage (including lobar, putaminal, thalamic, and caudate) and (2) infratentorial hemorrhage (including cerebellar, pontine, mesencephalic). American Heart Association Stroke Outcome Classication was used to assess the severity of impairment and functional outcome.19 The severity of impairment(s) was dened as the following: (1) level A, no/ minimal neurologic decit caused by stroke in any domain (neurologic domains are motor, sensory, vision, affect, cognition, and language functions); (2) level B, mild/moderate decit caused by stroke in more than 1 domain; and (3) level C, severe decit caused by stroke in more than 1 domain. The functional outcome of patients during follow-up was categorized as the following: (1)

152

E. KUMRAL ET AL.

independent, patient is able to live alone, maintain a household, and access the community for leisure or productive activities (requires minimal assistance); (2) partially dependent, patient is able to live alone with substantial daily help from family or community; (3) completely dependent, patient is unable to live alone safely and requires full-time care. Disability score after the stroke was quantied using Barthel index.20 The severity of functional impairment, functional outcome, and disability score of patients were assessed at 24 hours, 48 hours, and 72 hours on the 7th day and after 3 months of stroke onset. Statistical Methods Continuous variables are expressed as mean standard deviation or as median. Proportions were compared with chi square test in a univariate analysis for screening of variables. The time at which hyperthermia was recorded during the rst 72 hours for stroke prognosis was assessed by stepwise logistic regression analysis. Correlations between the highest body temperature (38C) recorded at every hour period and stroke subtypes, topography, and stroke outcome measure at 3 months were performed with Cox proportional hazards regression models. Cutoff points for BI was 1 (poor outcome, 0-50); 2 (moderate outcome, 51-75); and 3 (good outcome, 76-100). Functional outcome evaluation was assessed as independent, partially dependent, completely dependent and death, and severity of impairment was considered as level a, b, and c, with consideration of the mean values in hyperthermic and normothermic patients. Origins of hemorrhagic stroke were considered in 1 group because hypertension was the main origin of hemorrhagic stroke. Kaplan-Meier survival curves were obtained to describe 3-month mortality. All analyses were performed with the SPSS data analysis system (SPSSInc, Chicago, IL).

Table 1. Baseline characteristics of the 473 patients with acute stroke Ischemic stroke (n 387) Hemorrhagic stroke (n 86) 59.5 10 53/33 65 (76) 61 (71) 18 (21) 28 (33) 2 (2) 13 (15) 9 (11) 3 (4) 5 (6) 0 5 (6) 2 (2) 8 (9) 18 (21) 28 (33) 23 (27) 21 (24) 11 (13) 16 (19) 65 (76) 2 (2) 3 (4) 16 (19) 28 (33) 17 (20) 3 (4) 38 (44)

Variables

Results
There were 250 men (mean age, 61.7 11 years; range, 48-83), and 223 women (mean age, 65.4 10 years; range, 44-80 years). Three hundred-six patients had infarct in the anterior circulation, and 81 had infarct in the posterior circulation, whereas 68 patients had supratentorial hemorrhage, and 18 had infratentorial hemorrhage. The origin of ischemic stroke was LAA in 199 patients, CE in 73, SAD in 60, coexistence of LAA and cardioembolic in 31, and undetermined in 24 patients. The origin of hemorrhagic stroke was hypertension in 65 patients (76%). Table 1 shows the demographic data risk factors, and stroke subtypes. Among 473 patients, 53 patients (11%) had hyperthermia (38.2 1.4C) at hospital admission, and 82 patients (17%) showed hyperthermia (38.3 1.2C) during the rst 3 days of stroke onset. Sixteen

Age (y) (SD) 63.5 11 Gender (M/F) 197/190 Hypertension 151 (39) Current smoking 140 (36) Diabetes mellitus 70 (18) Hypercholesterolemia 134 (35) History of stroke 19 (5) History of MI 81 (21) Atrial brillation 77 (20) Hematocrit (50) 31 (8) History of migraine 21 (5) Carotid stenosis (70%) 25 (7) (30%-69%) 149 (39) Vertebral/basilar artery stenosis (30%) 25 (6) Hyperuricemia 34 (9) Hyperthermia (38C) At admission 35 (9) 4 hours 42 (11) 12 hours 56 (14) 24 hours 61 (16) 48 hours 43 (11) 72 hours 58 (15) Cause of ischemic stroke LAA 199 (51) CE 73 (19) SAD 60 (16) Mixed etiology (LAA CE) 31 (8) Undetermined 24 (6) Cause of hemorrhagic stroke Hypertension Vascular malformations Anticoagulation Undetermined Functional outcome Independent 171 (44) Partially dependent 117 (30) Dependent 19 (5) Death 80 (21)

NOTE. Values in parentheses are percentage. Abbreviations: CE, cardioembolic; MI, myocardial infarction.

percent (61/387) of patients with ischemic stroke and 24% (21/86) of those with hemorrhagic stroke presented with hyperthermia. The body temperature was higher in patients with LAA (OR 3.98, 95% CI 2.16-8.97, P .001), and hemorrhagic stroke (OR 2.05, 95% CI 1.07-8.68, P .001) than those with SAD (OR 0.5, 95% CI 0.001-0.63, P .5) and cardioembolic stroke (OR 1.27,

EFFECT OF ORIGIN AND TOPOGRAPHY

153

Figure 1. Kaplan-Meier survival curve showing the cumulative survival rate of patients with high body temperature ( 38C) and normal temperature. Patients with hyperthermia had worse survival rates than those without after 3 months of stroke onset (P .001). (- - -, hyperthermia; , normothermia.)

95% CI 0.93-1.18, P .40 ). In patients with posterior circulation infarct, the body temperature was higher than those with anterior circulation infarct (OR 3.71, 95% CI 2.07-6.67, P .001), whereas there was no difference between patients with infratentorial hemorrhage and those with supratentorial hemorrhage (OR 1.04, 95% CI 0.75-1.43, P .08). Three months after stroke, in logistic regression analysis, high body temperature at 24 hours (OR 2.17, 95% CI 2.09-7.57, P .001) and 48 hours (OR 1.27, 95% CI 1.06-4.84; P 0.02) was correlated with poor outcome and mortality (Fig 1). Mortality ratio was 63.4% (52/82) in patients with hyperthermia; it was 16.8% (66/ 391) in those with hypothermia (OR 2.27, 95% CI 1.70-3.03, P .001) (Table 2). BI scores at discharge were signicantly lower in hyperthermic patients (mean, 49.4 17.8) than normothermic patients (mean, 64.2 19.4) (P .001). The correlation coefcients between the mean body temperature recordings within 72 hours and BI scores decreased signicantly after 3 months of stroke onset (P .0001, r 0.26).

Table 2. Univariate Cox regression analysis of functional outcome regarding origins of stroke and topography of lesions in patients with hyperthermia vs normothermia Risk ratio (95% CI) Independent (n 199) 1.62 (0.98-4.26) 0.002 (0.001-0.54) 0.64 (0.35-1.16) 0.12 (0.10-4.92) 1.83 (0.44-7.41) Partially dependent (n 134) 0.33 (0.18-6.82) 0.66 (0.54-4.71) 1.34 (0.07-2.41) 0.66 (0.40-0.98) 2.81* (0.78-10.16) Completely dependent (n 22) 1.66* (1.01-10.12) 1.99 (0.61-6.40) 0.32 (0.11-9.44) 1.11 (0.18-2.88) 1.25 (0.34-4.55) Death (n 118) 0.38 (0.37-2.12) 0.61 (0.23-1.64) 0.37 (0.20-0.60)

Origins of stroke LAA (n 199) SAD (n 60) CE (n 73) Mixed Origin (LAA & CE) (n 31) Other and undetermined (n 24) Intracerebral hemorrhage (n 86) Topography of Lesions Total anterior circulation infarct (n 220) Partial anterior circulation infarct (n 51) Posterior circulation infarct (n 56) Supratentorial hemorrhage (n 68) Infratentorial hemorrhage (n 18)

0.25 (0.13-0.51) 0.22 (0.19-1.46) 0.54 (0.20-1.47) 0.57 (0.28-1.12)

0.52 (0.21-1.30) 0.72 (0.24-2.11) 2.59 (1.50-4.47) 1.94 (0.73-5.12)

0.57 (0.18-1.8) 0.61 (0.04-0.23) 1.67 (0.14-20.05)

0.54* (0.29-0.96) 2.97 (0.99-8.86) 2.20 (1.12-4.32) 0.64 (.23-1.81)

Abbreviation: CE, cardioembolism. *P .05. P .001. Not calculable because of lack of convergence in Cox regression.

154

E. KUMRAL ET AL.

Table 3. Univariate Cox regression analysis of severity of impairment(s) regarding origins of stroke and topography of lesions in patients with hyperthermia versus normothermia Risk ratio (95% CI) Level A (n 83) 1.33 (0.9-2.3) (0.001-0.54) 0.2 (0.22-2.92) 2.7 (0.89-5.56) 0.9 (0.5-1.8) 0.8 (0.4-1.4) 0.6 (0.3-1.4) 0.26 (0.35-2.8) Level B (n 282) 2.4 (1.1-3.8) (0.54-4.71) 0.14 (0.09-0.28) 2.8* (0.89-3.9) 1.7 (1.1-2.6) 1.8 (1.2-2.6) 0.7 (0.4-1.2) 0.4 (0.1-1.6) Level C (n 108) 1.7 (1.3-2.1) 1.9* (1.5-2.8) 2.3 (1.4-3.9) 3.0 (1.8-4.9) 1.9* (1.1-3.2) 4.5 (1.7-11.8)

Origins of stroke LAA (n 199) SAD (n 60) CE (n 73) Intracerebral hemorrhage (n 86) Topography of Lesions Total anterior circulation infarct (n 220) Partial anterior circulation infarct (n 51) Posterior circulation infarct (n 56) Supratentorial hemorrhage (n 68) Infratentorial hemorrhage (n 18)

Abbreviation: CE, cardioembolism. *P .05. P .001. Not calculable because of lack of convergence in Cox regression.

At 3 months after stroke, 83 patients (18%) had no/ minimal decit in any neurologic domain (level A), 35 of 247 (14%) patients with hyperthermia had mild decit in more than 1 neurologic domain (level B), and 47 of 61 (77%) hyperthermic patients had severe decit in more than 1 neurologic domain (level C). In logistic regression analysis, hyperthermic patients with LAA and cardioembolism had signicant severe neurologic decit in more than 1 neurologic domain compared with hypothermic patients (P .0001) (Table 3). Regarding lesion site in patients with total anterior circulation infarct (TACI), posterior circulation infarct (PCI), supra- and infratentorial hemorrhage, hyperthermia was associated with severe neurologic decit compared with hypothermia at 3 months after onset of stroke (P .001, P .0001, P .02, P .001, respectively).

Discussion
Our ndings suggest that body temperature in 3 days of acute stroke affect the prognosis and mortality depending on origin and localization of lesion. A similar trend for in-hospital poor prognosis was found in a recent meta-analysis that included patients with ischemic and hemorrhagic stroke.15 The cause of hyperthermia after stroke and the effect of stroke subtypes with pyrexia on the prognosis are not always evident. Among stroke subtypes, patients with intraventricular or subarachnoid hemorrhages and brain stem infarct seem to have greater central or neurogenic fever.4,8 Patients with subarachnoid hemorrhage were not included in the design of our study.

Another possible cause of pyrexia at stroke onset could be a superimposed infection,4,9,21 but in our study, we excluded any cause of sepsis by clinical and laboratory examinations. Moreover, coexistent infection within the rst 3 days of stroke was not independently associated with poor prognosis,9 and results from a previous study generally support that fever may be directly related to the size of the cerebral lesior.1 A previous study on human stroke showed that hyperthermia within 72 hours of stroke signicantly may increase mortality and hyperthermia within the rst 24 hours after stroke, which may cause signicantly greater morbidity.9 In our analysis, we used 72 hours fever data affecting the prognosis of patients regarding the origin and topography of lesions. We found that in patients with large supratentorial hemorrhage, TACI and PCI, coexistent pyrexia had a signicant effect on mortality. Moreover, hyperthermic patients with LAA and supratentorial hemorrhage had signicantly greater morbidity compared with those with normothermia. Recent studies on hyperthermia-related brain damage in acute stroke showed that there was a strong relationship between intensity of hyperthermia within the rst 24 hours and larger infarct volume, higher neurologic decit, and dependency at 3 months.4,7 In contrast, hypothermia ( 36.5C) in ischemic stroke patients was associated with a signicant reduction of in-hospital mortality and morbidity.10,22 The probable mechanism for the higher mortality and morbidity after hyperthermia in ischemic stroke seems to be related to the size of the area with functionally im-

EFFECT OF ORIGIN AND TOPOGRAPHY

155 2. Ginsberg MD, Strenau LL, Globus MY, et al. Therapeutic modulation of brain temperature: Relevance to ischaemic brain injury. Cerebrovasc Brain Metab Rev 1992; 4:189-225. 3. Azzimondi G, Bassein L, Nonino F, et al. Fever in acute stroke worsens prognosis. A prospective study. Stroke 1995;26:2040-2043. 4. Reith J, Jorgensen HS, Pedersen PM, et al. Body temperature in acute stroke: Relation to stroke severity, infarct size, mortality, and outcome. Lancet 1996;347:422-425. 5. Davalos A, Castillo J, Pumar JM, et al. Body temperature and brinogen are related to early neurological deterioration in acute ischemic stroke. Cerebrovasc Dis 1997;7: 64-69. 6. MacWalter R, McMahon A, Fraser H, et al. Does body temperature on admission predict long-term outcome after an acute stroke? Cerebrovasc Dis 1998;8:11 (suppl, abstr). 7. Castillo J, Davalos A, Marrugat J, et al. Timing for fever-related brain damage in acute ischemic stroke. Stroke 1998;29:2455-2460. 8. Albrecht RF, Wass CT, Lanier WLL. Occurrence of potentially detrimental temperature alterations in hospitalized patients at risk for brain injury. Mayo Clin Proc 1998;73:629-635. 9. Ginsberg M, Busto R. Combating hyperthermia in acute stroke: A signicant clinical concern. Stroke 1998;29:529534. 10. Wang Y, Lim LLY, Levi C, et al. Inuence of admission body temperature on stroke mortality. Stroke 2000;31: 404-409. 11. Busto R, Dietrich WD, Globus MY, et al. Small differences in intraischemic brain temperature critically determine the extent of ischemic injury. J Cereb Blood Flow Metab 1987;7:729-738. 12. Wass CT, Lanier WL, Hofer RE, et al. Temperature changes of 1C alter functional neurologic outcome and histopathology in a canine model of complete cerebral ischemia. Anesthesiology 1995;83:325-335. 13. Kim Y, Busto R, Dietrich WD, et al. Delayed postischemic hyperthermia in awake rats worsens the histopathological outcome of transient focal cerebral ischemia. Stroke 1996;27:2274-2281. 14. Kakuda W, Naritomi H, Shimizu T, et al. Body temperature increases following embolic stroke correlating with the size of infarction. J Cereb Blood Flow Metab 1997; 17:43 (suppl, abstr). 15. Hajat C, Hajat S, Sharma P. Effects of poststroke pyrexia on stroke outcome. Stroke 2000;31:410-414. 16. Svensson LG, Crawford ES, Hess KR, et al. Deep hypothermia with circulatory arrest. Determinants of stroke and early mortality in 656 patients. J Thorac Cardiovasc Surg 1993;106:19-31. 17. Crowder CM, Tempelhoff R, Theard MA, et al. Jugular bulb temperature: comparison with brain surface and core temperature in neurosurgical patients during mild hypothermia. J Neurosurg 1996;85:98-103. 18. Bamford J, Sandercock P, Dennis M, et al. Classication and natural history of clinically identiable subtypes of cerebral infarction. Lancet 1991;337:1521-1526. 19. American Heart Association Stroke Outcome Classication. Stroke 1998;29:1275-1280. 20. Wade DT, Mewer RL. Functional abilities after stroke:

paired but potentially viable tissue surrounding the irreversibly damaged tissue, called the ischemic penumbra. Hyperthermia causes the transformation of ischemic penumbra into infarction and accelerates the development of ischemic necrosis either by breaking the blood-brain barrier and changing cerebral blood ow at the vascular level23 or by specic alterations in neurotransmitter release,24 excitation of apoptosis, and increasing the inammatory response at the molecular level.25,26 In hemorrhagic stroke, MRI evidence points to another pathophysiologic process rather than penumbral region,27 a hypointense peripheral rim around the hematoma in the acute phase that represents either early phagocytic activity or a boundary region between blood and brain.28 The size of this critical zone in the large supratentorial hemorrhage could be a contributing factor, causing higher incidence of mortality and morbidity in patients with hyperthermia. The level of fever is also an important predictor of outcome in patients with stroke,3,7,10 but what the critical threshold of detrimental fever is is still debatable. Experimental studies provide evidence that ischemic neuronal injury may be increased signicantly with even mild hyperthermia of up to 2C above normal body temperature.3 We dened hyperthermia as body temperature above 38C as in most of the previous studies.3,7,21 According to this criteria, we found that among the origins, LAA, cardioembolism, and intracerebral hemorrhage may cause severe neurologic decits in more than 1 neurologic domain in the presence of hyperthermia compared with normothermia. In our study, topographic analysis showed that patients with TACI, PCI, and supraand infratentorial hemorrhage with hyperthermia at stroke onset seems to have detrimental effect in more than 1 neurologic functional domain after 3 months of stroke. In conclusion, our data suggest that hyperthermia with total anterior circulation infarct, posterior circulation infarct, and supratentorial hemorrhage is associated with a marked increase of 3-month mortality. Large artery disease, cardioembolism, and supra- and infratentorial hemorrhage associated with hyperthermia may increase the severity of neurologic decits. Particularly in the acute phase of stroke, therapeutic interventions against hyperthemia must remain one of the goals in the clinical setting to prevent stroke progression.

References
1. Przelomski MM, Roth RM, Gleckman RA, et al. Fever in the wake of a stroke. Neurology 1986;36:427-429.

156 Measurement, natural history and prognosis. J Neurol Neurosurg Psych 1987;50:177-182. Terent A, Andersson B. Prognosis for patients with cerebrovascular stroke and transient ischaemic attacks. Ups J Med Sci 981;86:63-74. Wass CT, Lanier WL. Hypothermia-associated protection from ischemic brain injury: implications for patient management. Int Anesthesiol Clin 1996;34:95-111. Dietrich WD, Halley M, Valdes I, et al. Interrelationships between increased vascular permeability and acute neuronal damage following temperature-controlled brain ischaemia in rats. Acta neuropathol (Berl) 1991;81:615625. Ginsberg MD, Busto R, Martinez E. The effects of cerebral ischaemia on energy metabolism. In: Schousboe A, Diemer NH, Kofod H, eds. Drug Research related to

E. KUMRAL ET AL. neuroactive Amino Acids: Alfred Benzon Symposium 32. Copenhagen, Denmark: Munksgaard; 1992:207-224. Maier CM, Ahem KVB, Cheng ML, et al. Optimal depth and duration of mild hypothermia in a focal model of transient cerebral ischemia. Effects on neurologic outcome, infarct size, apoptosis, and inammation. Stroke 1998;29:2171-2180. Boisvert D. In vivo assessment of hydroxyl radical free radical production in the brain. J Cereb Blood Flow Metab 1991;11:S637 (abstr). Hirano T, Read S, Abbott D, et al. No evidence of ischaemic penumbra after cerebral hemorrhge. Cerebrovasc Dis 1998;8:53 (suppl, abstr). Patel MR, Edelman RR, Warach S. Detection of hyperacute primary intraparenchymal hemorrhage by magnetic resonance imaging. Stroke 1996;27:2321-2324.

21.

25.

22.

26.

23.

27.

24.

28.