| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Department of Anesthesiology, Duke University Medical Center, Durham, North Carolina; and Department of Perioperative Care and Emergency Medicine, University Medical Center Utrecht, Utrecht, The Netherlands.
Address correspondence and reprint requests to Hilary P. Grocott, MD, Professor of Anesthesiology and Surgery, University of Manitoba, Adjunct Professor of Anesthesiology, Duke University, Asper Institute for Clinical Research, 369 Tache Ave., Room CR3008, Winnipeg, Manitoba, Canada R2H 2A6. Address e-mail to hgrocott{at}sbgh.mb.ca.
Abstract
BACKGROUND: Clinical studies have failed to demonstrate significant benefits of hypothermia for the prevention of postoperative cognitive dysfunction (POCD) after cardiopulmonary bypass (CPB). One explanation for this might be that potentially injurious cerebral hyperthermia occurs during rewarming at the end of CPB, off-setting the protective benefits of hypothermia. In this study, we investigated the relative influence of CPB temperature, rewarming strategies, and postoperative temperature in a rat CPB model.
METHODS: Four groups of male Sprague-Dawley rats were surgically prepared and subjected to 90 min of CPB. Group A was normothermic (37.5°C) during and after CPB. Group B underwent hypothermic (32°C) CPB, followed by rewarming to 37.5°C at the end of bypass. Group C had hypothermic (32°C) CPB, followed by limited rewarming to 35°C. Group D had normothermic CPB with hypothermia (35°C) induced only postoperatively. Groups were compared for POCD determined by the performance in the Morris water maze on postoperative days 3–9. Histologic analysis of the brains (CA1 and CA3 hippocampal regions) was also performed.
RESULTS: Hypothermia induced only during (group B versus group A) or after CPB (group D versus group A) conferred no significant POCD benefit. Hypothermia when induced during CPB and continued into the postoperative period resulted in a significant improvement in water maze performance versus all other temperature regimens (group C versus group A, P = 0.044; group C versus group B, P = 0.011; group C versus group D, P = 0.012). No histological differences among groups were demonstrated.
CONCLUSIONS: The combination of hypothermic (32°C) CPB coupled with limited rewarming and prolonged postoperative hypothermia (35°C) decreased POCD after CPB in rats.
Cardiac surgery using cardiopulmonary bypass (CPB) has been associated with a frequent incidence of postoperative cognitive dysfunction (POCD).1,2 The etiology of cognitive impairment after cardiac surgery is multifactorial and may include cerebral microembolization, cerebral hypoperfusion, systemic and cerebral inflammation, cerebral temperature perturbations, cerebral edema, and possible blood–brain barrier dysfunction, all superimposed on genetic influences that may alter susceptibility to injury or alter repair from injury once it has occurred.3 Many neuroprotective strategies, pharmaceutical as well as surgical, have been investigated, but clinical success has been limited.3–5 Extensive laboratory studies have demonstrated that hypothermia is a potent therapeutic tool to attenuate the effects of experimental cerebral ischemia.6–9 However, several large clinical studies in cardiac surgical patients have failed to demonstrate any benefits of hypothermia for the prevention of POCD.10–12 One possible explanation for this lack of effect may be related to the rewarming that occurs before the end of CPB. Overly aggressive rewarming can result in the brain being exposed to dangerously elevated blood temperatures that may neutralize any neuroprotective effect afforded by hypothermia and might itself be deleterious to the brain.13 One clinical trial did demonstrate a benefit of hypothermic CPB by integrating limited rewarming while maintaining prolonged postoperative hypothermia.14 However, a clear distinction whether the patients benefited from the postoperative hypothermia per se or the avoidance of rewarming could not be made. The purpose of this study was to investigate the effect of different CPB temperatures, rewarming strategies, and postoperative temperature on POCD after CPB in a rat model.
METHODS
The study was approved by the Duke University Animal Care and Use Committee and all procedures met the National Institutes of Health guidelines for animal care.15
Anesthesia, Surgical Preparation, and Cardiopulmonary Bypass
Male 375–400 g Sprague-Dawley rats (Harlan; Indianapolis, IN) were housed two per cage under 12-h light–dark cycle conditions with free access to food and water. After a 12-h fast, anesthesia was induced with 3% isoflurane in 50% O2 in a plastic box. After orotracheal intubation with a 14-gauge catheter (Insyte BD Medical, Sandy, UT), the animals were mechanically ventilated (Harvard Model 687, Harvard Apparatus, Holliston, MA) to maintain a normal arterial carbon dioxide tension. During the surgical preparation and during recovery, anesthesia was maintained with 1.5%–2.0% isoflurane.
Surgical preparation consisted of cannulation of the tail artery with a 20-gauge catheter (Insyte BD Medical, Sandy, UT), which later served as arterial inflow for the CPB circuit. One hundred and fifty international units of porcine heparin was given after placement of the first catheter. Arterial blood pressure was monitored via the superficial caudal epigastric artery, which was cannulated with polyethylene tubing (PE-10 Intramedic Tubing, Becton-Dickinson, Sparks, MD). A modified multiorifice 4.5 F pediatric catheter (modified Desilets-Hoffman Catheter, Cook, Bloomington, IN) was advanced through the right external jugular vein into the right heart for venous outflow. The CPB circuit16 consisted of a glass venous reservoir, a peristaltic pump (Masterflex©, Cole-Parmer Instrument CO, Vernon Hills, IL) and a membrane oxygenator (Micro© neonatal oxygenator, Cobe Cardiovascular, Arvada, CO), all of which were connected with 1.6 mm ID silicone tubing (Tygon©, Cole-Parmer Instrument CO, Vernon Hills, IL). An in-line flow probe (2N806 probe and T208 flowmeter, Transonics Systems Inc., Ithaca, NY) was used to continuously measure CPB flow.
The CPB circuit was primed with approximately 30 mL of fresh whole blood obtained from two heparinized (100 IU per animal) donor animals. As necessary, 6% hetastarch (Hextend, Hospira Inc., Lake Forest, IL) was added to the reservoir to maintain an adequate circulating volume. During CPB, ventilation was discontinued, and anesthesia was maintained with fentanyl (30 µg/kg IV), midazolam (0.4 mg/kg IV), and atracurium (0.5 mg/kg IV) as a bolus injection, followed by a continuous infusion of 2.5 µg · kg–1 · min–1 fentanyl, 0.03 mg · kg–1 · min–1 midazolam, and 0.08 mg · kg–1 · min–1 atracurium. The oxygenator fresh gas flow (4% CO2, 66% O2, and 30% air) was approximately 1 L/min.
All rats underwent 90 min of CPB and were randomized into four groups based on different perioperative and postoperative temperature regimens (n = 13 per group). Group A was subjected to 90 min of normothermic (37.5°C) CPB, followed by a 6-h postoperative normothermic period of 37.5°C. Group B underwent hypothermic CPB (32°C) for 75 min with rewarming over the last 15 min of CPB to 37.5°C. This temperature was then maintained for the first 6 h after CPB. Group C was subjected to 75 min of hypothermic CPB at 32°C with limited rewarming at the end of CPB to a temperature of 35°C, which was maintained for the first six postoperative hours. Group D was normothermic during CPB, cooled to 35°C at the end of CPB and then maintained at that temperature for the first six postoperative hours. After 6 h, all animals were allowed to rewarm to a normothermic temperature.
During surgery, CPB, and the first 6 h after CPB, rectal and pericranial temperature were measured and servo-regulated (YSI 400 series thermistor and 73ATA Indicating controller, YSI, Yellow Springs, OH) according to group designation. The heating/cooling system consisted of a heating blanket, a convective forced-air system, as well as a water-jacketed venous reservoir and arterial flow inlet in addition to the operating platform that was connected to either a warm water bath (36°C–38°C) or a water bath with ice water. At different time points (Table 1), blood gas analysis (
stat) and hemoglobin levels were performed using an IL1306 blood gas analyzer (Instrumentation Laboratories Inc., Lexington, MA) and an OSM3 Hemoximeter© (Radiometer Inc., Copenhagen, Denmark). The targeted flow during CPB was 160–180 mL · kg–1 · min–1 closely resembling the normal cardiac output in the rat.17 The animals were weaned from CPB without the use of any medication and heparin anticoagulation was allowed to dissipate spontaneously. The venous cannula and the epigastric arterial catheter were then removed and the wounds closed. The animals’ lungs were ventilated for 6 h after CPB to allow any remaining neuromuscular blockade to dissipate as well as to maintain temperature control. After the last blood sample, the tail cannula was removed, the wound closed, and anesthesia discontinued. After tracheal extubation, the animals were placed in an oxygen-enriched environment.
|
Cognitive Testing
Starting on the third postoperative day and continuing for 7 days, the animals underwent cognitive testing in the Morris water maze (MWM).16 The MWM is a 1.5 m diameter black pool filled with water (26°C ± 1°C) with a fixed submerged platform in one quadrant and various visual clues present on the surrounding walls. The time taken by the rat to locate the hidden platform after having been placed in the pool (defined as the latency) was measured. During the testing period, the animals underwent daily testing with four trials per day. At each trial, the animals were placed in the pool in randomly assigned, different quadrants. If the platform could not be found, the maximum time spent swimming in the pool was limited to 90 s, after which the animal was placed on the platform for 15 s. Their performance (including swimming speed) was recorded and analyzed using a computerized video tracking system (Ethovision, Noldus, Wageningen, the Netherlands).
Histologic Examination
After completion of the MWM testing, the animals were anesthetized with 3% isoflurane and underwent in situ brain fixation using an intracardiac injection of heparinized saline followed by 10% buffered formalin. Twenty-four hours later, the brains were removed and stored in 4% formalin. Paraffin-embedded sections were serially cut in 5-µm slices and then stained with hematoxylin and eosin for light microscopic evaluation. From each brain, the most dorsal slice with two cross-sections of each lateral ventricle (±3.6 mm from bregma) was examined, and the total number of necrotic neurons in bilateral hippocampal CA1 and CA3 regions were recorded. Cytoplasmatic eosinophilia, karyorrhexis, or pyknosis were used as criteria for cellular necrosis.
Statistical Analysis
MWM latencies were analyzed using repeated measurements analysis of variance with Bonferroni post hoc testing. Physiologic values and histological outcomes were compared using one-way analysis of variance with Bonferroni post hoc testing where appropriate. Statistical significance was assumed when P < 0.05.
RESULTS
There was one death in group A, three in group B, and three in group D, all related to surgical/technical difficulties, including excessive blood loss during cannulation or mechanical failure of the pump connections during the procedure. Thus, there were 45 rats in the experiment available for analysis (group A: n = 12; group B: n = 10; group C: n = 13; and group D: n = 10 rats).
Physiologic values at different time points during the experiment are presented in Table 1. The temperature differences at several time points were in accordance with the a priori group designations. There was a transiently higher arterial oxygen tension in group C, the group with hypothermia during CPB and in the postoperative period. All other variables were not significantly different among the groups.
Figure 1 is the average latency for the animals to find the platform. In all groups, there was daily improvement in the MWM latency, providing evidence of learning over time (P < 0.001) and a plateau was reached, implying no further learning. There were no differences in MWM latency between groups A versus B, A versus D, and B versus D (P = 1.00 for all comparisons). However, group C (hypothermic bypass with limited rewarming and prolonged postoperative hypothermia) performed better than all other groups (group C versus group A, P = 0.044; group C versus group B, P = 0.011; group C versus group D, P = 0.012). The swimming speeds averaged between 39 ± 23 and 44 ± 25 cm/s throughout the postoperative days with no significant differences among the four groups (P = 0.206) indicating no differences in motor capabilities among groups.
|
Histologic results are listed in Table 2. There were no differences among the four groups in the number of a dead hippocampal CA1 (P = 0.25), or for CA3 neurons (P = 0.46).
|
DISCUSSION
In the present study, we subjected rats to 90 min of CPB with different intraoperative and postoperative temperature regimens. The animals were subjected to either a normo- or hypothermic (32°C) CPB period in combination with a 6 h hypothermic (35°C) or normothermic (37.5°C) postoperative period. In comparison to the other groups, the hypothermic CPB group with a period of postoperative hypothermia performed significantly better on the MWM test, indicating better cognitive performance. This suggests that isolated hypothermia (during or after CPB) offers no significant cognitive protection in comparison to normothermic temperature strategies, but that a combined strategy (hypothermic CPB with limited rewarming and a prolonged hypothermic period) decreases CPB-related POCD in rats. We speculate that this was most likely due to the benefits of limited rewarming, as the group that only had postoperative hypothermia demonstrated no neuroprotective benefit.
This study was designed to answer a pertinent clinical question. Although hypothermia itself is found to be neuroprotective in different types of studies,6–9 in cardiac surgery, its neuroprotective properties remain unclear.10–12,18 Hypothermic bypass has been examined in a number of trials focusing on neurocognitive outcome. McLean et al.10 were unable to detect any neuroprotective effect from moderate hypothermia when compared to normothermia in a study of 201 patients. Similarly, Mora et al.18 randomly assigned patients to hypothermic (<28°C) or normothermic (>35°C) temperature regimens and also found no influence of CPB temperature strategy on neurocognitive performance. In that study, neurocognitive performance deteriorated in more than half of each treatment group, and at the 6-wk follow-up assessment, 15% in each group continued to have cognitive impairments. Importantly, they also reported a 10% incidence of new neurologic dysfunction (i.e., stroke) in the normothermic group versus none in the hypothermic group. In comparison, the Warm Heart Investigators trial, a study with 1732 patients undergoing bypass surgery,12 showed no difference in stroke rate between two similar temperature groups.
A clinical study from our own investigative group compared normothermic (35.5°–36.5°C; n = 136) CPB with hypothermic (28°–30°C; n = 134) CPB for change in cognitive function 6 wk postoperatively and found no difference between the two temperature groups.11 One possible explanation for the lack of difference in the various hypothermia versus normothermia studies is that all of the hypothermic patients eventually had to undergo rewarming. This rewarming itself may have negated any benefit conferred by hypothermia by causing an over-shoot in cerebral temperature.19 Another limitation of these studies was the definition of normothermia. Some allowed temperature drift such that the patients were actually mildly hypothermic; others actively rewarmed the patients such that potential harmful cerebral hyperthermia occurred.
To address the potential confounding effect of rewarming on neurocognitive outcome after hypothermic bypass, a second clinical trial was performed in which 165 patients undergoing hypothermic CPB (28°C–32°C) were assigned to a conventional or slow rewarming rate.13 In the slow rewarming group, a 2°C difference between nasopharyngeal-perfusate temperatures was maintained during rewarming, whereas in the conventional group, temperature gradients of 4°–6°C were allowed. A significant association between change in cognitive function and rate of rewarming was found, with the slow rewarming group performing significantly better with respect to cognitive outcome.
The results of the present experiment corroborate the finding that hypothermic CPB is beneficial when coupled with limited rewarming and prolonged postoperative hypothermia. A slow rewarming rate appears to be essential for a beneficial effect on cognitive outcome. In our study, the bypass inflow temperature was never allowed to increase beyond a normothermic (37.5°C) level. This prevented any over-shoot in cerebral temperature to be responsible for the poorer outcome in the group that was rewarmed before the end of CPB. Even in the absence of cerebral hyperthermia, the known uncoupling of cerebral blood flow and metabolism during rewarming20 may have still have led to ischemic injury in the brain.
The experimental design in this rat CPB model allowed for the investigation of different temperature strategies, especially those that cannot be performed in the clinical setting. For example, in group D (normothermic CPB and hypothermic postoperative period), it was relatively easy to perform and allowed the assessment of the influence of a single postoperative hypothermic period after a normothermic CPB period. In the clinical setting, this isolated postoperative hypothermia would be difficult to justify.
Nathan et al.14 conducted a study in which 223 patients undergoing coronary artery bypass graft were all cooled to 32°C at the start of CPB and randomly assigned to rewarming protocols with targets of either 34°C or 37°C. Surface rewarming was applied in the intensive care unit with strict avoidance of hyperthermia. One week after surgery, the hypothermic group had a lower incidence and severity of postoperative cognitive impairment, indicating possible beneficial effects of mild hypothermia. In a recent investigation, the same authors21 addressed the safety issues raised by mild postoperative hypothermia. They concluded that postoperative hypothermia is likely safe (with respect to bleeding outcomes, inotropic use, and times to tracheal extubation), and that complete rewarming after hypothermic CPB is not always necessary. In contemporary cardiac surgery, patients are not usually intentionally cooled without subsequent rewarming, but the Nathan et al. results as well as the present animal study suggest that this avoidance of rewarming may have some merit if combined with mild intraoperative hypothermia.
There were several limitations in this study. Unlike in the clinical setting, we did not measure perfusate-nasopharyngeal temperature gradients, as has been used as a means of regulating rewarming rate.11 However, we did measure pericranial temperature, a close correlate of brain temperature, by inserting a temperature probe between the temporal muscle and the cranium. In this way, we achieved cooling and rewarming within 15 min without any over-shoot in pericranial temperatures. A second limitation was the lack of any discernible histological correlate injury to the cognitive differences. This absence of neurotic injury is not completely unexpected, as brain dysfunction often occurs without any gross histological injury. This absence of necrotic brain injury in the presence of a functional deficit has been seen in models of CPB.16,22 A final limitation was our relatively short-term outcome. We chose to kill the animals after cognitive testing on day 9. This was done in an effort to balance the logistics of accommodating a reasonable postoperative period of neurocognitive testing with a window within which gross histopathological changes (though not seen here) might possibly be discernable.
In summary, we have demonstrated a neuroprotective effect of limited rewarming and prolonged postoperative hypothermia after hypothermic CPB in rats. This finding is consistent with clinical studies of various hypothermia and rewarming strategies and provides further evidence for limiting the rewarming rate and accepting some postoperative hypothermia.
Footnotes
Accepted for publication November 13, 2007.
Supported, in part, by a starter grant to Dr. Grocott from the Foundation for Anesthesia Education Research (FAER).
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  R Binnema, A van der Wal, C Visser, R Schepp, L Jekel, and P Schroder Treatment of accidental hypothermia with cardiopulmonary bypass: a case report Perfusion, May 1, 2008; 23(3): 193 - 196. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
