Stanford University Libraries

Transcription

1 Page 1 of 18 Stanford University Libraries Expanded Academic ASAP Plus Journal of Neuroscience Nursing, Feb 2004 v36 i1 p23(9) Comparison of brain temperature to core temperature: a review of the literature. Laura Mcilvoy. Full Text: COPYRIGHT 2004 American Association of Neuroscience Nurses Abstract: In both animal models and human studies examining acute neurological injury, elevated core temperatures have been shown to exacerbate the degree of neuronal injury. There is an assumption that core temperature and brain temperature are the same. With the introduction of brain temperature monitoring technology, it has become possible to examine the difference between core and brain temperatures. The purpose of this integrated review was to examine the published literature comparing core temperatures (blood, rectal, bladder, and esophageal) with brain temperatures (measured by direct contact with the brain or measured in any of the spaces surrounding the brain, excluding intraoperative measurements). Fifteen studies from 1990 and 2002 were found. All 15 studies found that brain temperature was higher than all measures of core temperature with mean differences of 0.39 to 2.5[degrees]C reported. Only three studies employed a t test to examine the differences; all found statistical significance. Temperatures greater than 38[degrees]C were found in 11 studies. This review demonstrates that brain temperatures have been found to be higher than core temperatures; however, existing studies are limited by low sample sizes, limited statistical analysis, and inconsistent measures of brain and core temperatures. Because fever is prevalent in acutely injured neurological patients, its detection and treatment are essential interventions. In the absence of brain temperature monitoring, detection of a 'brain fever' may be limited. Future research is needed to further examine the relationship between brain and core temperatures and their impact on intracranial dynamics. ********** Both animal models and human studies have overwhelmingly demonstrated that hyperthermia, when present during or after a period of brain injury or ischemia, exacerbates the degree of resulting neuronal injury (Azzimondi et al., 1995; Busto et al., 1987; Dietrich, 1992; Ginsberg & Busto, 1998; Hajat, Hajat, & Sharma, 2000; Reith et al., 1996; Wang, Lim, Levi, Heller, & Fisher, 2000). Since the inception of the technology to measure direct brain temperature, few studies have looked at the differences between brain temperatures and core temperatures. However, there is mounting evidence that brain temperatures are higher than core temperatures in patients with acute neurological injuries. If the temperature of an injured brain is higher than body temperature, episodes of neural hyperthermia may go undetected possibly leading to worsening of neurological injury. This article provides an integrated review of selected research that examines the differences between brain temperature and core temperature. Its purpose is to discuss the methodological challenges within these studies, evaluate the statistical and clinical significance of the findings, identify gaps, and make recommendations for future research. Selection Process

2 Page 2 of 18 A computerized search of MEDLINE and the Cumulative Index to Nursing and Allied Health was performed using the keywords brain and temperature to identify relevant literature since The reference lists of identified articles were searched for additional studies. The selection of studies included in this review was based on the following criteria: * The study compared a brain temperature, measured by direct contact with the brain or measured in any of the spaces surrounding the brain (epidural or subdural), with any type of core temperature measurement (blood temperature, rectal temperature, bladder temperature, or esophageal temperature) within a human population. * The brain temperature measurements were not collected intraoperatively with the dura open. Methodological Challenges of Brain Temperature Research Fifteen studies published between 1990 and 2002 examining the relationship between brain temperature and core temperature were found. Table 1 details the sample size, methods of brain and core temperature measurement, methods and statistical analysis used, and findings of the 15 studies. Sample sizes ranged from 6 to 63 with 9 studies having 20 or fewer subjects. Nine studies reported mean temperature only and made conclusions based upon the degree on which they differed. Only six studies utilized statistical analysis (Table 1). Temperature Sites Temperature within the brain varies by site and depth. Mellergard (1994, 1995) and Mellergard and Nordstrom (1990, 1991) found temperature measured in the epidural space was always lower than the temperature measured in the lateral ventricle by a gradient of [degrees]C. Hirashima and colleagues (1998) measured brain temperature at one cm intervals from the surface of the brain to the anterior horn of the lateral ventricle in patients with hydrocephalus (Hirashima et al., 1998). Temperature increased gradually with depth in all patients with a significant correlation between depth and brain temperature (r = 0.724, p <.0001). The highest temperature was found in the lateral ventricle (4-5 cm from the brain surface). Intraventricular brain temperature was significantly higher than rectal temperature, 37.4[degrees]C versus 36.7[degrees]C (p <.001, n = 21), whereas brain temperature at 2cm depth (the depth most intraparenchymal catheters are positioned) was significantly lower than rectal temperature, 36.2[degrees]C versus 36.7[degrees]C (p <.025, n = 21). Sternau and colleagues (1991) found ventricular brain temperature measured 3 cm distal to a cortical microthermistor were [degrees]C higher than the cortical temperature. Schwab discovered that cortical temperature exceeded epidural temperature by up to 2.0[degrees]C with a mean difference of 1.0[degrees]C (Schwab, Schwarz, Aschoff, Keller, & Hacke, 1998; Schwab, Spranger, Aschoff, Steiner, & Hacke, 1997). Upon examining offsets of mean differences of subdural space temperature and esophageal temperature, Mariak, Jadeszko, Lewko, Lewkowski, and Lyson (1998) reported that arterial blood seemed to remain cooler than brain surface and that temperature of brain parenchyma may be higher than that of its surface, irrespective of fever or normothermia. Of the nine studies published since 1997 examining the relationship between brain and core temperatures, five utilized epidural and/or intraparenchymal brain temperature readings. These studies comparing epidural and intraparenchymal temperatures with core temperature measurements may not be using the highest brain temperature. The gold standard of core temperatures would be arterial blood temperature as measured by a

3 Page 3 of 18 pulmonary artery catheter (O'Grady et al., 1998; Schellock & Rubin, 1982). However, to date only one published study has compared brain temperatures with pulmonary artery temperatures. The majority of studies examining brain temperature use rectal temperature as the measurement of core temperature. The literature is contradictory as to whether rectal temperature is an accurate reflection of pulmonary artery blood temperature. Some investigators report moderate to high correlations between pulmonary artery and rectal temperatures (r = 0.49 to r = 0.99; Fulbrook, 1993; Henker & Coyne, 1995; Klein et al., 1993; Lilly, Boland, & Zekan, 1980; Mravinac, Dracup, & Clochesy, 1989; Robinson, Charlton, Seal, Spady, & Joffres, 1998; Rotello, Crawford, & Terndrop, 1996; Schmitz, Bair, Falk, & Levine, 1995). Others report rectal temperatures consistently higher than pulmonary artery temperatures with mean differences of 0.04 to 0.5[degrees]C cited (Fulbrook; Henker & Coyne; Robinson et al.; Rotello et al.; Schmitz et al.). Studies utilizing brain temperature measurements from sites other than the ventricle may be reporting inaccurately low temperatures, whereas studies utilizing rectal temperatures as core temperatures may be reporting inaccurately high temperatures, artificially lowering mean differences. The difference between brain and core temperature may be higher than reported in the literature. Rectal Temperature Versus Brain Temperature Ventricular temperatures were found to be higher than rectal temperatures in all studies that compared the two variables with mean differences of [degrees]C reported (Henker, Brown, & Marion, 1998; Hirashima et al., 1998; Mariak, Jadeszko, Lewko, Lewkowski, & Lyson, 1998; Mellergard, 1994, 1995; Mellergard & Nordstrom, 1990, 1991; Schwab, Schwarz, et al., 1998; Schwab, Spranger et al., 1997; Soukup et al., 2002; Verlooy, Heytens, Veeckmans, & Selosse, 1995; Zauner et al., 1998). Rumana, Gopinath, Uzura, Valadka, and Robertson (1998) compared intraparenchymal and jugular bulb temperatures with rectal temperature. Intraparenchymal temperature was significantly higher than rectal temperature (p <.001), while there was no significant difference between jugular bulb temperature and rectal temperature. Bladder Temperature Versus Brain Temperature Because urine is a filtrate of blood, previous studies comparing bladder temperatures and pulmonary artery temperatures found high correlations (r = 0.78 to r = 0.94; Mravinac, Dracup, Clochesy, 1989) and low mean offset differences (0.03 [+ or -] 0.23[degrees]C; Erickson & Kirklin, 1993) between the two temperatures. Five of the reviewed studies used bladder temperature as a measure of core temperature. Brain temperature was higher than bladder temperature in all five studies by a gradient of [degrees]C reported (Henker et al., 1998; Hirashima et al., 1998; Mariak et al., 1998; Mellergard, 1994, 1995; Mellergard & Nordstrom, 1990, 1991; Schwab, Schwarz et al., 1998; Schwab, Spranger et al., 1997; Sternau et al., 1991; Verlooy et al., 1995). Bladder temperature was closer to brain temperature than rectal temperature in both studies that examined rectal and bladder core temperatures (Henker et al., 1998; Verlooy et al., 1995). Pulmonary Artery Temperature Versus Brain Temperature Only one study was found to examine the difference between brain temperature and pulmonary artery temperature (Rossi, Zanier, Mauri, Columbo, & Stocchetti, 2001). Brain temperature was measured in the lateral ventricle in 17 subjects and with an intraparenchymal probe in 3 subjects. Mean brain temperature was 0.3 [+ or -] 0.3[degrees]C higher than pulmonary artery temperatures with a range of -0.7 to 2.3[degrees]C. Seventy-three percent of brain temperatures and 57.5% of pulmonary artery

4 Page 4 of 18 temperatures were >38[degrees]C. The mean gradient between brain and pulmonary artery temperatures widened to 0.41 [+ or -] 0.38[degrees]C at the febrile peak (p <.05). Increases in brain temperature were associated with a significant increase in intracranial pressure that decreased as fever ebbed, from 17.5 [+ or -] 8.62 to 16 [+ or -] 7.76 mm Hg (p =.02). Isolated Findings Schwab et al. (1997) placed bilateral brain temperature monitors in seven patients experiencing acute middle cerebral artery (MCA) territory strokes. During the first 6 hours, the temperature in the infarcted hemisphere was 0.6[degrees]C higher than that in the contralateral hemisphere. Rumana et al. (1998) examined eight patients who were placed in barbiturate coma as a therapy for their intracranial hypertension. There was a trend for both intraparenchymal temperature and rectal temperature to decrease slightly within 5 hours after the loading dose, but the difference was not statistically significant. Summary of Brain Temperature Research All 15 studies found that brain temperature was higher than core temperature in the majority of participants (Table 1). The three studies that employed a t test to examine the differences found the difference of brain temperature higher than core temperature to be statistically significant (Hirashima et al., 1998; Rossi et al., 2001; Rumana et al., 1998). Temperatures greater than 38[degrees]C were found in 11 studies (Table 2). Mellergard and Nordstrom in their 1991 study found the majority of 15 subjects (exact number not given) had rectal temperatures greater than 38.0[degrees]C during some period of measurements. Three subjects had temperatures greater than 39.0[degrees]C for short periods, whereas no subjects had suspected infections or were treated with antibiotics. In 1994, Mellergard reported a mean ventricular brain temperature of 38.2[degrees]C in subjects with Reaction Level scores of 3-5 and 6-8. (Reaction Level scales are measures of level of consciousness similar to Glasgow Coma Scale scores.) Eight of 15 acute MCA territory stroke participants had bladder temperatures greater than 39.0 [degrees]c in the Schwab et al study. After treatment with antipyretics, mean core temperatures decreased by 1.1[degrees]C, while mean brain temperature decreased only 0.6[degrees]C. At 3 hours after treatment brain temperature had increased to previous values while core temperature took 5 hours to return to previous values. Henker et al. (1998) found seven of eight traumatic brain injury (TBI) subjects had temperatures greater than 38[degrees]C. Temperature differences between brain and bladder were more than 1.0[degrees]C in 53% of measurements in four of the seven subjects (Henker et al.). Seventy-three percent of brain temperatures and 57.5% of core temperatures were greater than 38.5[degrees]C as reported by Rossi et al. (2001) in an acute neurosurgical population. Soukup and colleagues (2002) recorded 3,979 observations of brain temperatures over 38.2 [degrees] 0.5[degrees]C In 58 TBI patients. Twenty-three of 63 subjects had temperatures greater than 38[degrees]C in Marick et al.'s 1998 study. Fourteen of these had brain temperatures higher than core temperature. This study calculated the differences between trunk temperature (rectal), esophageal temperature, and intracranial temperature in subjects with increasing fever, investigating the existence of a process of selective brain cooling. A significant reduction of these differences in step with increasing fever would be compatible with selective brain cooling. Their findings suggest that brain temperature in fever is not selectively suppressed.

5 Page 5 of 18 Rumana and colleagues (1998) found that the average brain temperature during the first five days after TBI was 38.9 [+ or -] 1.0[degrees]C, while the average rectal temperature was 37.8 [+ or -] 0.4 [degrees]c. It is notable that a temperature of 37.8[degrees]C is rarely considered febrile, while a temperature of 38.9[degrees]C is generally treated with cooling therapies. In the absence of brain temperature monitoring, this population appears to be afebrile and would not receive appropriate therapy. Discussion There is significant evidence supporting the hypothesis that brain temperatures are higher than core temperatures. This is important when considering the adverse outcomes associated with elevated core temperatures and its high prevalence. Fever and Outcome Elevated body temperatures have been shown to worsen outcomes in patients with acute neurological injuries. The adverse stroke outcomes of increased stroke severity, infarct size, morbidity, and mortality in ischemic stroke patients has been significantly associated with the presence of pyrexia (Azzimondi et al., 1995; Castillo, Davalos, Marrugat, & Noya, 1998; Grau et al., 1999; Hajat et al., 2000; Jorgensen, Reith, Pedersen, Nakayama, & Olsen, 1996; Reith et al., 1996; Terent & Andersson, 1981). In 2000, Wang and colleagues (2000) found a hyperthermic admission temperature was associated with an increase in mortality at 1 year, while Boysen and Christenson (2001) found a correlation between higher body temperature at 8 hours post stroke and poor outcome. The Copenhagen Stroke Study examined 725 consecutive stroke patients and found the expected correlation between high body temperature at 8 hours after stroke and poor outcome (Boysen & Christenson). A study examining nontraumatic subarachnoid hemorrhage patients found patients with symptomatic vasospasm have an increased risk of developing fever independent of disease severity or presence of infection (Ollveira-Filho et al., 2001). They also reported that there was an increased risk of poor outcome for each day of fever independent of disease severity, vasospasm, or infection. DeGeorgia, Charles, and Andresfsky (2001) imaged 61 patients with intracerebral hemorrhage and found fever at 72 hours after stroke was associated with higher mortality and worse outcome at 3 months. They also stated that a shift of the third ventricle may predict which stroke patients will develop fever in the next 72 hours. Another study within the stroke population examined the risk factors of pyrexia and dysphagia on mortality at 90 days (Sharma, Fletcher, Vassallo, & Ross, 2001). They found both risk factors independently and significantly associated with stroke severity and 90 day mortality. However, only dysphagia predicted mortality. Unlike stroke, there are few human studies examining outcome of fever in the TBI population. Natale, Joseph, Helfaer, and Shaffner (2000) examined 117 children with TBI admitted to a pediatric intensive care unit (PICU). Within the first 24 hours, 29.9% had a temperature greater than 38.5 [degrees]c. This early hyperthermia was an independent predictor of lower Glasgow Coma Scale score at PICU discharge and a longer PICU length of stay. Pyrexia was found to be among the most significant predictors of mortality in 124 adult TBI patients along with hypotension and hypoxemia (Jones et al., 1994). In examining 840 severe TBI, Jiang found that 25% experienced fevers greater than 39[degrees]C within 48 hours of injury. Of these, 39% died, 2% were vegetative, 27% had moderate to severe deficits, and only 24% reported good recovery (Jiang, Gao, Li, Yu, & Zhu, 2002).

6 Page 6 of 18 Fever Prevalence Because outcome studies demonstrate the detrimental effect of elevated temperature on injured brains, it is important to examine fever prevalence. Table 3 details sample sizes, methods of temperature measurement, fever definitions, and finding of studies that reported fever prevalence. In examining fever prevalence in the stroke population, Oliveira-Filho and colleagues (2001) reported 41% of 92 subarachnoid hemorrhage patients experienced fever, while Georgilis, Plomaritoglou, Dafni, Bassiakos, & Vemmos (1999) found a fever prevalence of 37.6% of 330 stroke patients. However, neither investigator specified the time after injury in which the fever occurred. Fever was found to occur within 48 hours in 12% of 725 stroke patients, and within 72 hours of the bleed in 42% of 250 intracerebral hemorrhage patients (Boysen & Christensen, 2001; Schwarz, Hafner, Aschoff, & Schwab, 2000). Fever with a plateau between 38[degrees]C and 39[degrees]C has been reported by day 5 in 88.3% of 107 patients experiencing arterial aneurysms with severe angiographic vasospasms (Rousseaux, Scherpereel, Bernard, Graftieaux, & Guyot, 1980). Studies examining fever in the mixed population (ischemic stroke, hemorrhagic stroke, and TBI) found in a neuroscience intensive care unit (ICU) reported fever rates of 16% within 24 hours of admission, 31.7% at hours after admission, 42% at 72 hours after admission, and 60%-70% at 48 hours to 96 hours after admission (Albrecht, Wass, & Lanier, 1998; Kilpatrick, Lowry, Firlik, Yonas, & Marion, 2000; Marion, 2001; Schwarz et at., 2000). Kilpatrick and colleagues (2000) reported fever rates of 93% for patients having an ICU length of stay greater than 14 days (Kilpatrick et al., 2000). Only one study was found that examined children, finding 29.9% of 17 children with TBI experienced fever within 24 hours of admission (Natale et al., 2000). Fever was prevalent in the brain temperature research studies as demonstrated by the eleven studies reporting temperatures [greater than or equal to] 38[degrees]C in the majority of their subjects. These studies clearly demonstrate that fever is prevalent early in patients with neurological injury. Implications * Fever is prevalent in acutely injured neurological patients. * In the absence of brain temperature monitoring, fever may be underdiagnosed and untreated as demonstrated by Rumana et al.'s 1998 study of TBI subjects. The technology of brain temperature monitoring has become very accessible, especially in systems that monitor ventricular pressure while providing CSF drainage. This review demonstrates the need for concurrent brain temperature monitoring with core temperature monitoring in order to discover and treat all fevers in acutely injured neurological patients. * The assumption that core temperatures and body temperatures are equal is false. All published studies have demonstrated that brain temperature is higher than core temperature. In fact, mean differences between brain and core temperatures may be higher than reported due to the use of less than gold standard temperature sites in the measurement of brain and core temperatures that are reported in the literature.

7 Page 7 of 18 Recommendations for Future Research To date, studies that have examined the relationship between brain and core temperatures are plagued by the problems of small sample sizes, limited statistical analysis, and inadequate measures of brain and core temperatures. Research is needed that includes adequate sample sizes and examines the relationship between ventricular brain temperatures and pulmonary artery core temperatures, within the neurologically impaired populations of stroke (both ischemic and hemorrhagic) and TBI. Discovering the relationship between brain and core temperatures across diagnostic groups will enable clinicians to make effective treatment decisions, especially in the absence of brain temperature data. Summary Both animal and human studies have demonstrated that elevated body temperature worsens the degree of injury produced by primary and secondary injury processes in the acutely injured neurological population. Considering the prevalence of fever in this population, methods to prevent and treat fever must be aggressively sought. Before the inception of the technology to measure direct brain temperature, it had been assumed that core temperature was an accurate reflection of brain temperature. If the temperature of an injured brain is higher than core temperature, episodes of neural hyperthermia may go undetected, possibly leading to further neurological injury. This review has demonstrated that brain temperature is predominantly higher than core temperature and that without monitoring of brain temperature fever detection may be limited. Table 1. Research Studies on Brain Temperature Method Authors Sample Size Brain Temp Core Temp Mellergard et Posterior fos- 7 Epidural, Rectal, al., 1990 sa; tumor (1); ventricular tympanic SAH (2); TBI (4) Sternau et al., TBI; posthem- 9 Ventricular, Bladder orrhagic cortical infarct; obstructive hydrocephalus Mellergard et Hydrocephalus 15 Ventricular Rectal al., 1991 (1); tumor (3); ICH (3); SAH (4); TBI (4) Mellergard, Neurosurgery 27 Epidural Rectal 1994 patients (10), ventricular (27) Mellergard, Neurosurgery 28 Epidural, Rectal 1995 patients ventricular Verlooy et al., TBI 6 Ventricular Rectal, 1995 bladder Schwab et al., Acute MCA; 15 Intraparen- Bladder 1997 territory stroke chymal (12),

8 Page 8 of 18 epidural (3) Henker et al., TBI: three nor- 8 Ventricular Rectal, 1998 mothermic; bladder five hypothermic Mariak et al., Neurosurgical 63 Intraparen- Rectal, 1998 procedures chymal (16), tympanic, subdural esophageal Hirashima et Hydrocephalus 21 Ventricular, Rectal al., 1998 with varying subdural etiologies Schwab et al., Acute MCA; 20 Intraparen- Bladder 1998 territory stroke chymal, epidural Rumana et al., TBI 30 Intraparen- Rectal 1998 chymal, jugular bulb (21) Rossi, et al., TBI, SAH, 20 Ventricular Pulmonary 2001 tumor (17); intra- artery parenchymal (3); jugular (15) Soukup et al., TBI 58 Ventricular, Rectal 2002 intraparenchymal Zauner et al., TBI 60 Ventricular, Rectal 1998 intraparenchymal Statistical Authors Analysis Methods Mellergard et Mean difference; Obtained epidural al., 1990 no SD brain temperature; then positioned probe in ventricle for 1-5 days Sternau et al., Mean difference; Ventricular, cortical, no SD bladder and intracranial pressure hourly readings Mellergard et Mean difference [+ or -] Ventricular and rectal al., 1991 SD temperatures from 8 hours to 7 days Mellergard, Mean difference [+ or -] Temperature difference 1994 SEM and Reaction Level Scale (RLS) scores Mellergard, 1995 Verlooy et al., Mean difference [+ or -] Ventricular, rectal and 1995 SD bladder temps from hours

9 Page 9 of 18 Schwab et al., Mean difference [+ or -] Intraparenchymal, 1997 SD epidural, and bladder monitored 3-7 days; 7 patients had bilateral monitors Henker et al., Mean difference [+ or -] Examined 3 ranges: 1998 SD ventricular temp [less than or equal to] 36[degrees]C; ventricular temp >36[degrees]C to [less than or equal to] 38[degrees]C; ventricular temp >38[degrees]C Mariak et al., Mean difference Retrospective analysis 1998 between trunk and over 4 years subdural cerebral temperature, and intraparenchymal linear smoothing with- (16 cases) temps, recleast squares to visu- tal used as core alize the regression, correlation coefficients Hirashima et Mean difference [+ or -] Brain temperature al., 1998 SD; paired nest; Spear- recorded 2 cm (intraman's rank correlation; parenchymal) depth Wilcoxon's U test and 4 cm (ventricular) depth Schwab et al., 1998 Rumana et al., Mean difference [+ or -] Intraparenchymal, 1998 SEM; repeated ANO- jugular bulb, and rec- VA; paired t test with tal hourly readings for Bonferroni correction hours for multiple comparisons; regression analysis Rossi, et al., Mean difference [+ or -] Ventricular, intra SD; paired t test parenchymal, pulmonary artery monitored hours Soukup et al., Pearson's coefficient; Examined 4 groups by 2002 mean temperature dif- mean brain temp; norference [+ or -] SD; mal temp linear regression 36-37[degrees]C; analysis hyperthermic temp >37.5[degrees]C with therapeutic cooling; temp <36[degrees]C (instituted with progressive ICP increase); spontaneous hypothermia

10 Page 10 of 18 temp <36[degrees]C Zauner et al., 1998 Authors Mellergard et al., 1990 Sternau et al., Mellergard et al., 1991 Findings Mean ventricular brain temperature higher than rectal core temperature in 6 of 7 patients; temperature gradients 0.4-1[degrees]C between ventricular brain temperature and epidural brain temperature. Ventricular microthermistor recorded [degrees]C higher temperature than cortical microthermistor. Ventricular and core temperature generally 0.5[degrees]C higher than bladder temperature in 3 obstructive hydrocephalic subjects. Brain temperature [degrees]C higher than bladder temperature in 5 TBI subjects. Ventricular higher than rectal 90% measurements; mean temperature difference 0.37[degrees]C for all patients. Mellergard, Epidural always lower than ventricular; ventricular 1994 higher than rectal 90% of time; RLS 1-2, 0.4 [+ or -].07[degrees]C; RLS 3-5, 0.3 [+ or -].04 [degrees]c; RLS 6-8, 0.27 [+ or -].11[degrees]C Mellergard, Same data as in Mellergard 1994 study, included 1995 one more subject. Verlooy et al., Rectal curve deviated from brain temperature more 1995 markedly than bladder temperature brain higher than bladder in 5 patients 0.5 [+ or -] 0.2[degrees]C Schwab et al., Intraparenchymal higher than bladder 1.5 [+ or -] [degrees]C ventricular higher than epidural 10[degrees]C; ventricular higher than bladder 1.9[degrees]C; Infarcted hemisphere higher than contralateral hemisphere at 6 hours post infarct by 0.6[degrees]C. Henker et al., Difference between ventricular and bladder greater 1998 in [less than or equal to] 36[degrees]C and >38[degrees]C; 53% of measurements had a difference of >1[degrees]C in 4 of 7 patients in >38[degrees]C; 43% of measurements had a difference of >1[degrees]C in 3 of 7 patients in [less than or equal to] 36[degrees]C; difference ventricular and bladder 0.3 [+ or -] 0.25[degrees]C to 1.9 [+ or -] 1.52[degrees]C, difference ventricular and rectal 1.32 [+ or -] 0.32[degrees]C to 2 [+ or -] 1.08[degrees]C. Mariak et al., The offsets rectal-subdural, rectal-intraparenchy mal, and esophageal-subdural were plotted against rectal over a wide range of body temperature with near zero correlation found. Hirashima et al., 1998 Significant correlation between depth and temperature r = 0.724; p =.001; ventricular mean 37.4 [+ or -] 0.83[degrees]C; rectal mean 36.7 [+ or -] 0.7 [degrees] C, p =.001; intraparenchmal

11 Page 11 of 18 mean 36.2 [+ or -] 0.95 [degrees] C; rectal mean 36.7 [+ or -] 0.7 [degrees] C, p =.025 Schwab et al., Same data as in Schwab 1997 study, includes dead subjects and includes induced hypothermia Rumana et al., Mean intraparenchymal 38.9 [+ or -] 1.0[degrees]C 1998 and mean rectal 37.8 [+ or -].4 [degrees] C, p <.001; jugular bulb mean (n = 14) 37.7 [+ or -] 0.5 [degrees] C, p =.63; age of patient was inversely related to difference between brain and rectal temperature, [r.sup.2] =.14, p =.038; cerebral metabolic rate for oxygen and cerebral metabolic rate for glucose were not significantly related to brain temperature (n = 16). Rossi, et al., Mean brain temperature 38.4 [+ or -] 0.8[degrees]C; 2001 mean core temperature 38.1 [+ or -] 0.8[degrees]C; mean difference 0.3 [+ or -] 0.3[degrees]C (p = ). In 12% of subjects, core temperature > brain temperature; increases in brain temperature were associated with a significant rise in ICP from 14.9 [+ or -] 7.9[degrees]C to 22 [+ or -] 10.4[degrees]C (p <.05). As fever decreased, there was a significant decrease in ICP from 17.5 [+ or -] 8.62 to 16 [+ or -] 7.76 mm Hg (p =.02). Mean gradient between brain and core temperature was 0.16 [+ or -] 0.31[degrees]C before febrile episode and 0.41 [+ or -] 0.38[degrees]C at febrile peak (p < 0.05) with brain temperature higher than core temperature. Soukup et al., significant correlation between brain and rectal 2002 temperature (r = 0.866). Linear regression analysis showed adjusted [R.sup.2] of Differences between brain and rectal temperatures (brain-rectal) showed brain temperature higher than rectal temperature in normothermic and hyperthermic groups. Normothermic group (0.0 [+ or -] 0.5[degrees]C), hyperthermic group (0.3 [+ or -] 0.5[degrees]C), 38.2 [+ or -] 0.5[degrees]C brain; 37.9 [+ or -] 0.5[degrees]C rectal; therapeutic cooling (-0.2 [+ or -] 0.6[degrees]C); hypothermia (-0.8 [+ or -] 1.4[degrees]C). During first 24 hours after injury 67% had brain temperature higher than 37.5[degrees]C with temperature difference of 0.4 [+ or -] 0.6[degrees]C. Zauner et al., Data same as in Soukup 2002 study above, more subjects added. Emphasis of study on brain oxygen levels during variable levels of oxygen delivery. Table 2. Presence of Fever in Research Studies on Brain Temperature Author Fever Definition Findings Mellergard et Fever = rectal core Majority of 15 subjects had al., 1991 temperature [greater fever during some period of than or equal to] measurement; 3 subjects 38.0[degrees]C with temperature >39.0[degrees]C for short periods; no subjects with suspected infection; no antibiotics used.

12 Page 12 of 18 Mellergard, No definition of fever Mean ventricular 1994, 1995 temperature 38.2[degrees]C in subjects with Reaction Level Scale scores 3-5 and 6-8. Schwab et al., No definition of fever 12 of 15 subjects had 1997, 1998 bladder temperature above 39.0[degrees]C during measurement period. Henker et No definition of fever 53% of measurements had a al., 1998 but designed study with difference of 3 different temperature >1.0[degrees]C in ranges, the highest one >3.08[degrees]C range in 4 >38.0[degrees]C of 7 patients. Mariak et Fever = rectal core 23 of 63 patients had al., 1998 temperature [greater fever; 14 febrile patients than or equal to] had brain temperature 38.0[degrees]C higher than rectal core temperature. Rumana et No definition of fever Average brain temperature al., 1998 first 5 days after TBI in 30 subjects 38.9 [+ or -] 1[degrees]C; average rectal core temperature 37.8 [+ or -] 0.4[degrees]C. Rossi et Fever = temperature 73% of brain temperature al., 2001 [greater than or equal measurements [greater to] 38.0[degrees]C than or equal to] 38.0[degrees]C; mean brain temperature 38.4[degrees]C; 57.5% of core temperature measurements [greater than or equal to] 38.0[degrees]C; mean core temperature 38.1[degrees]C. Soukup et No definition of fever 67% of patients showed a al., 2002; brain temperature higher Zauner et than 37.5[degrees]C during al., 1998 the first 24 hours after injury; brain temperatures of 38.2 [+ or -] 0.5[degrees]C were recorded in 3,979 recorded observations. Table 3. Fever Prevalence in Research Studies on Brain Temperature Temperature Authors Sample Method Rousseaux et 107 arterial aneurysms Rectal al., 1980 Albrecht et 40 subarachnoid Unknown

13 Page 13 of 18 al., 1998 hemorrhage; 40 traumatic brain injury; 40 postcardiac arrest Georgilis et 330 stroke Unknown al., 1999 Natale et 117 pediatric traumatic Unknown al., 2000 brain injury Schwarz et 251 intensive care unit Oral or rectal al., 2000 Kilpatrick et 428 neuro intensive Rectal al., 2000 care unit Marion, neuro intensive care unit Boysen et 725 acute stroke Tympanic al., 2001 Oliveira-Filho 92 subarachnoid Tympanic et al., 2001 hemorrhage Fever Authors Definition Findings Rousseaux et Between Fever observed in 88.3% of al., [degrees]C patients by day 5 and lasting on and 39[degrees]C average 9 days. Albrecht et >38[degrees]C 73.3% experienced fever; al., % of traumatic brain injury, 24% of cardiac arrest, and 4% of subarachnoid hemorrhage experienced fever [greater than or equal to] 39[degrees]C. Georgilis et None 37.6% had fever. al., 1999 Natale et >38.5[degrees]C 29.9% had fever within first 24 al., 2000 hours of admission. Schwarz et [greater than Admission temperatures: 18% al., 2000 or equal to] [greater than or equal to] 38.5[degrees]C 37.5[degrees]C; 1% >38.5[degrees]C; 198 patients followed for 72 hours, 42% had at least one episode of temperature [greater than or equal to] 38.5[degrees]C. Kilpatrick et >38.5[degrees]C 46.7% had fever; intensive care al., 2000 unit stay <24 hours had 16% febrile incidence; intensive care unit stay >14 days had 93% febrile incidence; 31.7% of patients had at least one febrile episode during intensive care unit stay of hours, 60%-70% of patients had at least one febrile episode during intensive care unit stay of 48 to 96 hours.

14 Page 14 of 18 Marion, 2001 Same data as in study above. Boysen et None 5.3% had temperature al., 2001 >37.5[degrees]C on admission; 12% had temperature >38[degrees]C in first 48 hours. Oliveira-Filho >38.3[degrees]C 41 % had fever for at least 2 et al., 2001 consecutive days. Acknowledgments This article was supported by a grant from Integra NeuroSciences, Plainsboro, NJ. References Albrecht, R., Wass, T., & Lanier, W. (1998). Occurrence of potentially detrimental temperature alterations in hospitalized at risk for brain injury. Mayo Clinical Proceedings, 73, Azzimondi, G., Bassein, L., Nonino, F., Fiorani, L., Vignatelli, L., Re, G., et al. (1995). Fever in acute stroke worsens prognosis. Stroke, 26, Boysen, G., & Christensen, H. (2001). Stroke severity determines body temperature in acute stroke. Stroke, 32, Busto, R., Dietrich, W., Globus, M.-T., Valdes, I., Scheinberg, P., & Ginsberg, M. (1987). Small differences in intraischemic brain temperature critically determine the extent of ischemic neuronal injury. Journal of Cerebral Blood Flow Metabolism, 7, Castillo, J., Davalos, A., Marrugat, J., & Noya, M. (1998). Timing for fever-related brain damage in acute ischemic stroke. Stroke. 29, DeGeorgia, M., Charles, B., & Andresfsky, J. (2001). Fever is associated with third ventricular shift in intracerebral hemorrhage. Paper presented at the 26th International Stroke Conference, Fort Lauderdale, FL. Dietrich, W. (1992). The importance of brain temperature in cerebral injury. Journal of Nearotrauma. 9, S475-S485.

15 Page 15 of 18 Erickson, R. S., & Kirklin, S. (1993). Comparison of ear-based, bladder, oral, and axillary methods for core temperature measurement. Critical Care Medicine, 21(10), Fulbrook, P. (1993). Core temperature measurement: A comparison of rectal, axillary and pulmonary artery blood temperature. Intensive, & Critical Care Nursing, 9(4), Georgilis, K., Plomaritoglou, A., Dafni, U., Bassiakos, Y., & Vemmos, K. (1999). Aetiology of fever in patients with acute stroke. Journal of Internal Medicine, 246, Ginsberg, M., & Busto, R. (1998). Combating hyperthermia in acute stroke: A significant clinical concern. Stroke, 29, Grau, A., Buggle, F., Schnitzler, P., Spiel, M., Lichy, C., & Hacke, W. (1999). Fever and infection early after ischemic stroke. Journal of the Neurological Sciences, 171, Hajat, C., Hajat, S., & Sharma, P. (2000). Effects of poststroke pyrexia on stroke outcomes: A metaanalysis of studies in patients. Stroke, 31(2), Henker, R., & Coyne, C. (1995). Comparison of peripheral temperature measurements with core temperature. AACN Clinical Issues, 6(1), Henker, R. A., Brown, S. D., & Marion, D. W. (1998). Comparison of brain temperature with bladder and rectal temperatures in adults with severe head injury. Neurosurgery, 42(5), Hirashima, Y., Takaba, M., Endo, S., Hayashi, N., Yamashita, K., & Takaku, A. (1998). Intracerebral temperature in patients with hydrocephalus of varying aetiology. Journal of Neurology, Neurosurgery, and Psychiatry, 64, Jiang, J., Gao, G., Li, W., Yu, M., & Zhu, C. (2002). Early indicators of prognosis in 846 cases of severe traumatic brain injury. Journal of Neurotrauma, 19, Jones, R, Andrews, R, Midgley, S., Anderson, S., Piper, I., Tocher, J., et al. (1994). Measuring the burden of secondary insults in head injured patients during intensive care. Journal of Neurosurgical Anesthesiology, 6(1), Jorgensen, H., Reith, J., Pedersen, P., Nakayama, H., & Olsen, T. (1996). Body temperature and outcome in stroke patients. The Lancet, 348, 193. Kilpatrick, M., Lowry, D., Firlik, A., Yonas, H., & Marion, D. (2000). Hyperthermia in the neurosurgical intensive care unit. Neurosurgery, 47(4), Klein, D., Mitchell, C., Petrinec, A., Monroe, M., Oblak, M., Ross, B., et al. (1993). A comparison of pulmonary artery, rectal, and tympanic membrane temperature measurement in the ICU. Heart & Lung, 22(5), Lilly, J., Boland, J., & Zekan, S. (1980). Urinary bladder temperature monitoring: A new index of body core temperature. Critical Care Medicine, 8(12), Mariak, Z., Jadeszko, M., Lewko, J., Lewkowski, W., & Lyson, T. (1998). No specific brain

16 Page 16 of 18 protection against thermal stress in fever. Acta Neurochirurgica, 140, Marion, D. (2001). Therapeutic moderate hypothermia and fever. Current Pharmaceutical Design, 7, Mellergard, P. (1994). Monitoring of rectal, epidural, and intraventricular temperature in neurosurgical patients. Acta Neurochirurgica, 60, Mellergard, P. (1995). Intracerebral temperature in neurosurgical patients: Intracerebral temperature gradients and relationships to consciousness level. Surgical Neurology, 43, Mellergard, P., & Nordstrom, C.H. (1990). Epidural temperature and possible intracerebral temperature gradients in man. British Journal of Neurosurgery, 4, Mellergard, P., & Nordstrom, C.H. (1991). Intracerebral temperature in neurosurgical patients. Neurosurgery, 28(5), Mravinac, C., Dracup, K., & Clochesy, J. (1989). Urinary bladder and rectal temperature monitoring during clinical hypothermia. Nursing Research, 38(2), Natale, J., Joseph, J., Helfaer, M., & Shaffner, D. (2000). Early hyperthermia after traumatic brain injury in children: Risk factors, influence on length of stay and effect on short-term neurologic status. Critical Care Medicine, 28, O'Grady, N., Barie, P., Bartlett, J., Bleck, T., Garvey, G., Jacobi, J., et al. (1998). Practice parameters for evaluating new fever in critically ill adult patients. Critical Care Medicine, 26(2), Oliveira-Filho, J., Ezzeddine, M. A., Segal, A. Z., Buonanno, F. S., Chang, Y., Ogilvy, C. S., et al. (2001). Fever in subarachnoid hemorrhage: Relationship to vasospasm and outcome. Neurology, 56, Reith, J., Jorgensen, H., Pedersen, R, Nakayama, H., Raaschou, H., Jeppesen, L., et al. (1996). Body temperature in acute stroke: Relation to stroke severity, infarct size, mortality, and outcome. The Lancet, 347, Robinson, J., Charlton, J., Seal, R., Spady, D., & Joffres, M. (1998). Oesophageal, rectal axillary, tympanic and pulmonary artery temperatures during cardiac surgery. Canadian Journal of Anaesthesia, 45(4), Rossi, S., Zanier, E., Mauri, I., Columbo, A., & Stocchetti, N. (2001). Brain temperature, core temperature, and intracranial pressure in acute cerebral damage. Journal of Neurology, Neurosurgery, & Psychiatry, 71(4), Rotello, L., Crawford, L., & Terndrop, T. (1996). Comparison of infrared ear thermometer derived and equilibrated rectal temperatures in estimating pulmonary artery temperatures. Critical Care Medicine, 24(9), Rousseaux, P., Scherpereel, B., Bernard, M., Graftieaux, J., & Guyot, J. (1980). Fever and cerebral vasospasm in ruptured intracranial aneurysms. Surgical Neurology, 14(6),

17 Page 17 of 18 Rumana, C. S., Gopinath, S. P., Uzura, M., Valadka, A. B., & Robertson, C. S. (1998). Brain temperature exceeds systemic temperature in head-injured patients. Critical Care Medicine, 26(3), Schellock, F., & Rubin, S. (1982). Simplified and highly accurate core temperature measurements. Medical Progress Technology, 8, Schmitz, T., Bair, N., Falk, M., & Levine, C. (1995). A comparison of five methods of temperature measurement in febrile intensive care patients. American Journal of Critical Care, 4(4), Schwab, S., Schwarz, S., Aschoff, A., Keller, E., & Hacke, W. (1998). Moderate hypothermia and brain temperature in patients with severe middle cerebral artery infarction. Acta Neurochirurgica, 71, Schwab, S., Spranger, M., Aschoff, A., Steiner, T., & Hacke, W. (1997). Brain temperature monitoring and modulation in patients with severe MCA infarction. Neurology, 48, Schwarz, S., Hafner, K., Aschoff, A., & Schwab, S. (2000). Incidence and prognostic significance of fever following intracerebral hemorrhage. Neurology, 54, Sharma, J., Fletcher, S., Vassallo, M., & Ross, I. (2001). What influcences outcome of stroke--pyrexia or dysphagia? International Journal of Clincal Practice, 55(1), Soukup, J., Zauner, A., Doppenberg, E., Menzel, M., Gilman, C., Young, H., et al. (2002). The importance of brain temperature in patients after severe head injury: Relationship to intracranial pressure, cerebral perfusion pressure, cerebral blood flow, and outcome. Journal of Neurotrauma, 19 (5), Sternau, L., Thompson, C., Dietrich, W., Busto, R., Globus, M., & Ginsberg, M. (1991). Intracranial temperature observation in the human brain. Journal of Cerebral Blood Flow and Metabolism, 11, S123. Terent, A., & Andersson, B. (1981). The prognosis for patients with cerebrovascular stroke and transient ischemic attacks. Upsala Journal of Medical Sciences, 86(1), Verlooy, J., Heytens, L., Veeckmans, G., & Selosse, P. (1995). Intracerebral temperature monitoring in severely head injured patients. Acta Neurochirurgica, 134, Wang, Y., Lim, L., Levi, C., Heller, R., & Fisher, J. (2000). Influence of admission body temperature on stroke mortality. Stroke, 31, Zauner, A., Doppenberg, E., Soukup, J., Menzel, M., Young, H., & Bullock, R. (1998). Extended neuromonitoring: New therapeutic opportunities? Neurological Research, 20(Suppl. 1), S85-S90. Questions or comments about this article may be directed to: Laura Mcilvoy, PhD RN CCRN CNRN, by at lmcilvoy@att.net. She is a doctoral candidate at Indiana University School of Nursing. Article A

18 Page 18 of 18 About Expanded Academic ASAP Copyright and Terms of Use