Stress-induced fever after postischemic rectal temperature measurements in the gerbil

Temperature Control in Rodent Neuroprotection Studies: Methods and Challenges

Extensive animal research facilitated the clinical translation of therapeutic hypothermia for cardiac arrest in adults and hypoxic-ischemic injury in infants. Similarly, clinical interest in hypothermia for other brain injuries, such as stroke, has been greatly supported by positive findings in preclinical work. The reliability, validity, and utility of animal models, among many research practices (blinding, randomization, etc.), are key to successful clinical translation. Here, we review methods used to induce and maintain hypothermia in animal models. These include physical and pharmacological methods. We emphasize the advantages and limitations of each approach, and the importance of using clinically relevant cooling protocols and appropriate monitoring and reporting approaches. Moreover, we performed a literature survey of ischemic stroke studies published in 2015 to highlight the continuing risk of temperature confounds in neuroprotection studies. For example, many still do not accurately monitor and report temperature during surgery (23.5%), even though almost half of these studies (46.0%) use pharmaceutical agents that likely influence temperature. We hope this review stimulates awareness and discussion of the importance of temperature in neuroprotective studies.

Incomplete Assessment of Experimental Cytoprotectants in Rodent Ischemia Studies

Background:

Inadequate preclinical testing (e.g., rodent studies) has been partly blamed for the failure of many cytoprotectants to effectively treat stroke in humans. For example, some drugs went to clinical trial without rigorous functional and histological assessment over long survival times. In this study, we characterized recent experimental practices in rodent cytoprotection experiments to determine whether the limitations of early studies have been rectified.

Methods:

We identified 138 rodent cytoprotection studies published in several leading journals (Journal of Neuroscience, Stroke, Journal of Cerebral Blood Flow and Metabolism and Experimental Neurology) for 2000 - 2002 and compared these to those published in 1990. From each study we determined the ischemia model, age and sex of the animal, the histological and functional endpoints used, and the methodology used to assess intra- and postischemic temperature.

Results:

Ninety-eight percent of recent studies used young adult rodents and most used males. Most studies (60%) did not assess functional outcome and survival times were often ≤ 48 hr (66%) for focal ischemia and ≤ 7 days (80%) for global ischemia. Over 60% of the experiments relied solely upon rectal temperature during ischemia and only 32.6% of ischemia studies measured temperature after surgery. The 1990 data were similar.

Conclusion:

Many investigators ignore the need to assess long-term functional and histological outcome and do not accurately represent clinical conditions of ischemia (e.g., use of aged animals). In addition, intra- and postischemic temperature measurement and control is frequently neglected or inadequately performed. Further clinical failures are likely.

Protein-energy malnutrition induces an aberrant acute-phase response and modifies the circadian rhythm of core temperature

Protein-energy malnutrition (PEM), present in 12%-19% of stroke patients upon hospital admission, appears to be a detrimental comorbidity factor that impairs functional outcome, but the mechanisms are not fully elucidated. Because ischemic brain injury is highly temperature-sensitive, the objectives of this study were to investigate whether PEM causes sustained changes in temperature that are associated with an inflammatory response. Activity levels were recorded as a possible explanation for the immediate elevation in temperature upon introduction to a low protein diet. Male, Sprague-Dawley rats (7 weeks old) were fed a control diet (18% protein) or a low protein diet (PEM, 2% protein) for either 7 or 28 days. Continuous core temperature recordings from bioelectrical sensor transmitters demonstrated a rapid increase in temperature amplitude, sustained over 28 days, in response to a low protein diet. Daily mean temperature rose transiently by day 2 (p = 0.01), falling to normal by day 4 (p = 0.08), after which mean temperature continually declined as malnutrition progressed. There were no alterations in activity mean (p = 0.3) or amplitude (p = 0.2) that were associated with the early rise in mean temperature. Increased serum alpha-2-macroglobulin (p < 0.001) and decreased serum albumin (p ≤ 0.005) combined with a decrease in serum alpha-1-acid glycoprotein (p < 0.001) suggest an atypical acute-phase response. In contrast, a low protein diet had no effect on the signaling pathway of the pro-inflammatory transcription factor, NFκB, in the hippocampus. In conclusion, PEM induces an aberrant and sustained acute-phase response coupled with long-lasting effects on body temperature.

Stress-induced rise in body temperature is repeatable in free-ranging Eastern chipmunks (Tamias striatus)

In response to handling or other acute stressors, most mammals, including humans, experience a temporary rise in body temperature (T(b)). Although this stress-induced rise in T(b) has been extensively studied on model organisms under controlled environments, individual variation in this interesting phenomenon has not been examined in the field. We investigated the stress-induced rise in T(b) in free-ranging eastern chipmunks (Tamias striatus) to determine first if it is repeatable. We predicted that the stress-induced rise in T(b) should be positively correlated to factors affecting heat production and heat dissipation, including ambient temperature (T(a)), body mass (M(b)), and field metabolic rate (FMR). Over two summers, we recorded both T(b) within the first minute of handling time (T(b1)) and after 5 min of handling time (T(b5)) 294 times on 140 individuals. The mean ∆T(b) (T(b5) - T(b1)) during this short interval was 0.30 ± 0.02°C, confirming that the stress-induced rise in T(b) occurs in chipmunks. Consistent differences among individuals accounted for 40% of the total variation in ∆T(b) (i.e. the stress-induced rise in T(b) is significantly repeatable). We also found that the stress-induced rise in T(b) was positively correlated to T(a), M(b), and mass-adjusted FMR. These results confirm that individuals consistently differ in their expression of the stress-induced rise in T(b) and that the extent of its expression is affected by factors related to heat production and dissipation. We highlight some research constraints and opportunities related to the integration of this laboratory paradigm into physiological and evolutionary ecology.

[Effect of social interaction on skin temperature in mice]

We investigated physiological and behavioral characteristics of socially stressed animals in a resident-intruder paradigm. ICR male mice (resident, n = 14) were exposed individually to a novel male conspecific (intruder, n = 14) in their homecage for 30 min. Along with behavioral analyses, the skin temperatures of both the resident and the intruder were measured simultaneously using a multipoint radiation thermometer. There were no significant differences between the resident and intruder in the amount of locomotion, flight and aggressive behaviors. The mean skin temperature of the residents during the interaction was higher than before the interaction. In addition, the skin temperatures of the intruders were consistently higher than the residents. The results suggest that social stress causes elevation in skin temperature as well as stress-induced hyperthermia in core temperature. Moreover, infrared radiation thermometers may provide an alternative means of measuring physiological parameters of two (or more) subjects simultaneously in the study of animal social behavior.

Delayed recovery and exaggerated infarct size by post-lesion stress in a rat model of focal cerebral stroke

Stress might be one of the most salient intrinsic factors influencing the risk of stroke and its outcome. Previous studies have linked stress to increased infarct size and exaggerated cognitive deficits in rodent models of stroke. This study compares the effects of chronic restraint stress, representing a psychological stressor, prior to or after motor cortex devascularization lesion on motor recovery in rats. Daily testing in a skilled reaching task revealed initially exaggerated deficits in limb use caused by pre-lesion stress in the absence of increased infarct size. Both pre- and post-lesion stresses affected movement by delaying recovery and limiting compensation of lesion-induced deficits. Nevertheless, only rats that experienced post-lesion stress showed enlarged infarct size. This was accompanied by enlarged edema formation in the lesion hemisphere of post-stress animals on day 2 post-lesion. There were no significant differences in infarct size between post-lesion day 2 and day 15. The data demonstrate that both pre- and post-lesion chronic restraint stresses affect motor recovery after ischemic lesion. Lesion volume, however, is influenced by the timing of a stressful experience relative to the lesion. These findings suggest that stress represents a critical variable determining the outcome after stroke.

Spontaneous temperature changes in the 2-vessel occlusion model of cerebral ischemia in rats

Transient global ischemia (ISC) in rats and humans causes selective and delayed neuronal death in the hippocampal CA1 sector. It is clear from rodent studies that hyperthermia aggravates, whereas hypothermia lessens, this injury. In this study we sought to relate core (Tc) and brain (Tb) temperature, measured via telemetry probes, after ISC produced in rats by bilateral common carotid artery occlusion combined with systemic hypotension (2-VO model). We also tested whether spontaneous postischemic temperature fluctuations occurred and whether they were related to cell death as previous studies indicate. We report that Tc and Tb readings are similar and are highly correlated before and after 10 min of 2-VO ISC. In the second experiment, rats were subjected to 8, 9, or 10 min of 2-VO ISC. Despite a range in CA1 injury among these animals, there was no evidence of post-ISC hyperthermia, contrary to earlier work, and neither temperature nor the physiological variables measured during ISC (e.g., glucose) predicted injury. Our findings suggest that, under the present conditions, 2-VO rats do not experience postoperative hyperthermia, which can be adequately measured with Tc telemetry probes.

Procedure of rectal temperature measurement affects brain, muscle, skin, and body temperatures and modulates the effects of intravenous cocaine

Rectal probe thermometry is commonly used to measure body core temperature in rodents because of its ease of use. Although previous studies suggest that rectal measurement is stressful and results in long-lasting elevations in body temperatures, we evaluated how this procedure affects brain, muscle, skin, and core temperatures measured with chronically implanted thermocouple electrodes in rats. Our data suggest that the procedure of rectal measurement results in powerful locomotor activation, rapid and strong increases in brain, muscle, and deep body temperatures, as well as a biphasic, down-up fluctuation in skin temperature, matching the response pattern observed during tail-pinch, a representative stressful procedure. This response, moreover, did not habituate after repeated day-to-day testing. Repeated rectal probe insertions also modified temperature responses induced by intravenous cocaine. Under quiet resting conditions, cocaine moderately increased brain, muscle, and deep body temperatures. However, during repeated rectal measurements, which increased temperatures, cocaine induced both hyperthermic and hypothermic responses. Direct comparisons revealed that body temperatures measured by a rectal probe are typically lower (approximately 0.6 degrees C) and more variable than body temperatures recorded by chronically implanted electrodes; the difference is smaller at low and greater at high basal temperatures. Because of this difference and temperature increases induced by the rectal probe per se, cocaine had no significant effect on rectal temperatures compared to control animals exposed to repeated rectal probes. Therefore, although rectal temperature measurements provide a decent correlation with directly measured deep body temperatures, the arousing influence of this procedure may drastically modulate the effects of other arousing stimuli and drugs.

Immediate constraint-induced movement therapy causes local hyperthermia that exacerbates cerebral cortical injury in rats

Constraint-induced movement therapy (CIMT), which involves restraint of the nonimpaired arm coupled with physiotherapy for the impaired arm, lessens impairment and disability in stroke patients. Surprisingly, immediate ipsilateral forelimb immobilization exacerbates brain injury in rats. We tested whether immediate ipsilateral restraint for 7 days aggravates injury after a devascularization lesion in rats. Furthermore, we hypothesized that ipsilateral restraint aggravates injury by causing hyperthermia. In experiment 1, each rat received two lesions, one in the motor cortex and one in the visual cortex. Ipsilateral restraint increased only the motor cortex lesion. In additional rats, no differences in core temperature occurred after ipsilateral or contralateral restraint. Thus, ipsilateral restraint does not aggravate injury by a systemic side effect. In experiment 2, we hypothesized that ipsilateral restraint causes hyperthermia in the region surrounding the initial cortical lesion. Brain temperature, measured via telemetry, was significantly higher (approximately 1 degrees C for 24 h) with ipsilateral restraint. A third experiment similarly found that ipsilateral restraint aggravates injury and causes local cortical hyperthermia and that contralateral restraint with externally induced mild hyperthermia aggravates injury. In conclusion, immediate ipsilateral restraint aggravates injury apparently by localized events that include hyperthermia. Caution must be exercised in applying early CIMT to humans, as hyperthermia is detrimental.

Post-ischemic diazepam does not reduce hippocampal CA1 injury and does not improve hypothermic neuroprotection after forebrain ischemia in gerbils

The hippocampal CA1 sector is especially vulnerable to brief forebrain ischemia. Excitotoxicity is widely thought to contribute to this cell death. Accordingly, drugs that presumably counteract excitotoxicity, such as GABAergic agonists, have been repeatedly tested and found to reduce CA1 cell loss. Post-ischemic diazepam reduces CA1 injury. However, diazepam also causes hypothermia, which by itself is neuroprotective. Most studies fail to adequately control for this confound. In this study, we tested whether diazepam reduces injury in temperature controlled gerbils subjected to brief forebrain ischemia. Furthermore, we tested whether diazepam augments hypothermic neuroprotection. All gerbils were implanted with a core temperature telemetry probe and a cannula for the subsequent insertion of a thermocouple probe to measure ischemic brain temperature. Subsequently, they were given a 5-min normothermic ischemic insult. In Experiment 1, two groups of gerbils were given 10 mg/kg doses of diazepam (i.p.) at both 30 and 90 min post-ischemia. Temperature was maintained in one group by heating lamps. Another group was administered saline. Diazepam reduced cell death at 7 days post-ischemia when the drug-induced hypothermia was permitted, but not when it was prevented. In Experiment 2, four groups of ischemic gerbils were treated starting at 12 h post-ischemia with prolonged hypothermia, diazepam and the combination or saline treatment. Hypothermia, but not diazepam, provided partial neuroprotection and diazepam did not augment hypothermic neuroprotection. Thus, neuroprotection with diazepam is solely due to hypothermia. These data do not support the clinical use of diazepam as a neuroprotectant after global ischemia.

Temperature-regulated model of focal ischemia in the mouse: a study with histopathological and behavioral outcomes

BACKGROUND AND PURPOSE

The importance of mouse stroke models has increased with the development of genetically manipulated animals. We hypothesized that immediate postischemia hypothermia may attenuate ischemic brain injury in the mouse.

METHODS

Intraabdominal radio frequency probes were implanted in animals and core temperature monitored. Groups included: MCAO-45-REG (45 minutes middle cerebral artery occlusion [MCAO]) temperature-controlled in the postischemic period >34 degrees C for 24 hours; MCAO-45 (45 minutes MCAO) were allowed to self-regulate core temperature during recovery; MCAO-30-REG (30 minutes MCAO), with the same temperature protocol as MCAO-45-REG; MCAO-30 (30 minutes MCAO), with temperature protocol the same as MCAO-45. Behavior and histological score was assessed at 7 days. The qualitative histological score assessed for injury in 18 specified regions.

RESULTS

MCAO-45-REG core temperature (mean 34.94 degrees C+/-0.8 degrees C) was significantly different than the self-regulating (MCAO-45, mean 33.1 degrees C+/-2.3 degrees C) for the first 4 hours after anesthesia (P<0.01). There was a trend toward similar differences in temperatures for MCAO-30-REG and MCAO-30 (P=0.08). At 7 days, a greater improvement in behavior score was observed for MCAO-45 and MCAO-30 compared with MCAO-45-REG and MCAO-30-REG (P<0.001). The histological score confirmed reduced injury in unregulated temperature groups (MCAO-45-REG mean 38+/-10 and MCAO-45 30+/-5.1, P<0.05; MCAO-30-REG 41+/-10 and MCAO-30 30+/-9, P<0.05).

CONCLUSIONS

Hypothermia is an important confounder of stroke injury in the intraluminal filament mouse model. Future mouse stroke studies must use strict temperature regulation.