Clinical studies have implicated surgery in promoting infections and compromising immune functions, including natural killer cell activity. Animal studies indicate that surgery-induced suppression of natural killer cell activity also promotes tumor metastasis. Hypothermia, a common surgical complication, has been suggested to underlie some of the deleterious consequences of surgery. This study evaluated the effect of hypothermia on the activity and number of blood natural killer cells and on host susceptibility to metastasis. The involvement of adrenergic mechanisms was also considered.


Fischer-344 rats remained awake in their cages (control group) or were anesthetized with 70 mg/kg thiopental and maintained for 2.5 h at core body temperatures of 30-32 degrees C (hypothermia group) or 38 degrees C (normothermia group). Thereafter, at several time points, blood was drawn so natural killer cell activity could be assessed, or rats were injected with syngeneic MADB106 tumor cells that metastasize only to the lungs. Lungs were removed 9 h later for assessment of lung tumor retention, or 4 weeks later for counting of metastases.


Normothermic anesthesia reduced natural killer cell activity (lytic units at 30% specific killing, mean +/- SEM) to 39+/-6.2% of control levels and hypothermia further reduced it to 15+/-6.6%. These changes were not accompanied by alterations in the numbers of circulating natural killer cells. Hypothermia increased tumor retention to 250% of control levels, and the number of metastases increased from 1.1+/-0.4 to 4.7+/-1.2. Normothermia had no significant effects on this index. Nadolol (0.4 mg/kg), a beta-adrenergic antagonist, significantly attenuated the effect of hypothermia on tumor retention.


Hypothermia under thiopental anesthesia suppresses natural killer cell activity and compromises host resistance to metastatic formation, possibly via adrenergic mechanisms. Such suppression may place patients with metastasizing tumors or dormant viral infections at greater risk for complications after intraoperative hypothermia.

IN humans, anesthesia and surgical procedures have been reported to suppress natural killer cell activity (NKCA) and other immune functions for several days. 1–4The biologic significance of such immune suppression is suggested by the well-established relation between surgical procedures and the outbreak of dormant infectious diseases known to be under immune surveillance (e.g. , tuberculosis, herpes, and cytomegalovirus). 5,6Whether such immunosuppressive effects of surgery also accelerate tumor development in the clinical setting is largely unknown and clearly difficult to evaluate, but animal studies have provided ample evidence for this possibility. 3,4,7 

Among immune mechanisms that control viral infection and malignancies are natural killer (NK) cells, a subpopulation of lymphocytes known to spontaneously recognize and kill various virally infected and tumor cells in vitro . 8–10Animal studies have shown that NK cells play an important role in controlling the development of leukemia and solid tumors, 10,11especially during the metastatic process. 12–16Human studies have indicated that levels of NKCA are negatively correlated with susceptibility to several types of cancer. 10Furthermore, higher levels of NKCA at the time of tumor removal have been associated with a better prognosis after excision of breast, 17head, or neck tumors. 18 

Intraoperative hypothermia occurs, at least to some extent, in more than 50% of surgical procedures. 19–21Intraoperative and exposure-related hypothermia have been shown to suppress several immune measures 22–24and to increase the risk of infections. 21,25,26Hypothermia elicits various physiologic responses, including sympathetic discharge and adrenal release of catecholamines and, to a lesser degree, glucocorticoids. 27,28These neuroendocrine responses have been reported to suppress immune functions, specifically NKCA. 29–31 

Because neuroendocrine responses affect the activity of individual NK cells and the distribution of NK cells among different immune compartments, 29,32–34both indices were assessed in the current study. To assess the health implications of potential changes in NKCA, we also used a model of metastatic formation based on the MADB106 tumor line, which is known to be sensitive to NK cells. This mammary adenocarcinoma cell line is syngeneic to the inbred Fischer-344 rats used in this study. After intravenous inoculation, MADB106 cells extravasate and reside in the lungs, and surviving cells develop colonies only in this organ. 12–14Studies have shown that these two related outcomes, lung tumor cell retention and the number of lung tumor colonies, are controlled by NK cells, primarily during the first 24 h after tumor injection. 12–14Therefore, in addition to providing a measure of host susceptibility to metastasis, the retention and colonization of MADB106 cells in the lungs provide an indication of NK cell performance in the whole animal.

In the current study, using a rat model, we sought (1) to determine whether hypothermia under barbiturate anesthesia suppresses the activity of NK cells or affects their distribution, (2) to determine whether hypothermia facilitates metastasis of an NK-sensitive malignancy, and (3) using a β-adrenergic antagonist, to identify the extent to which these effects are mediated by adrenergic mechanisms.


Fischer-344 male rats (Harlan Laboratories, Jerusalem, Israel) were housed four in a cage with free access to food and water. Animals were acclimatized to the vivarium for at least 4 weeks before the beginning of experiments and were 13–19 weeks old at that time. In any given experiment, all animals were of the same age. All experiments were approved by the Institutional Animal Care and Use Committee of Tel Aviv University.

Anesthesia, Hypothermia, and Normothermia 

Anesthesia was induced by intraperitoneal injection of 70 mg/kg sodium thiopental (Abbott S.p.A, Milan, Italy). Hypothermia was induced by positioning anesthetized rats on cotton-covered ice packs. Normothermia was maintained at 38°C (the normal rat body temperature) by positioning anesthetized rats on electric heating pads. Core body temperature was monitored using a 4-cm rectal probe (400 series; YSI Inc., Yellow Springs, OH). Anesthesia lasted for approximately 6 h in the normothermia group and 7 h in the hypothermia group.

Flow Cytometric Analysis 

Fluorescence-activated cell sorter analysis was used to determine the number of NK cells in the blood. These cells were identified as NKR-P1brightlymphocytes. 35To determine the total number of NK cells per milliliter of blood, a fixed number of polystyrene microbeads (20 μm; Duke Scientific, Palo Alto, CA) were added to the blood samples before they were prepared for cytometric analysis. For further details, see Shakhar et al.  29 

Whole-blood Natural Killer Cell Activity Assay 

The activity of NK cells was assessed using the whole-blood cytotoxicity assay described in detail elsewhere. 29Briefly, this procedure analyzes antitumor cytotoxicity of NK cells per milliliter of blood without previous purification of peripheral-blood mononuclear cells. One milliliter of blood is drawn by cardiac puncture, plasma is replaced with culture medium, and aliquots of the washed blood are placed in a microtiter plate. To assess NK cytotoxicity at six different effector-to-target ratios, six concentrations of radiolabeled YAC-1 tumor cells were added to the blood. Spontaneous and maximal releases of radioactivity from target cells are determined by substituting blood with the culture medium or hydrochloric acid, respectively. After a 4-h incubation period, samples of supernatant are recovered from each well so radioactivity could be assessed and specific target cell lysis could be calculated. Earlier studies indicated that cytotoxicity measured using this procedure is attributable to NK cells rather than to other cell types or soluble factors. 36,37 

The MADB106 Tumor Line 

MADB106 is a selected variant cell line obtained from a pulmonary metastasis of a mammary adenocarcinoma (MADB100) chemically induced in the inbred Fischer-344 rat. 14MADB106 cells were maintained in complete medium at 100% humidity, 5% carbon dioxide at 37°C, and detached from the flask using 0.25% trypsin.

Radiolabeling of Tumor Cells and Assessment of Lung Tumor Retention 

To assess lung tumor retention, DNA radiolabeling of tumor cells was accomplished by the addition of 0.4 mCi/ml of 125iododeoxyuridine (ICN Radiochemicals, Irvine, CA) to the growing cell culture 1 day before the cells were harvested for injection.

For tumor cell injection, rats were lightly anesthetized with halothane (Trofield Surgicals, London, UK), and 4 × 105/kg 125iododeoxyuridine-labeled MADB106 tumor cells in approximately 0.5 ml phosphate-buffered saline (supplemented with 0.1% bovine serum albumin) were injected into their tail veins. Nine hours later, rats were killed with halothane, and their lungs were removed and placed in a gamma counter for assessment of radioactive content. The percentage of tumor cells retained was calculated as the ratio of radioactivity measured in the lungs to total radioactivity in the injected tumor cells. Our previous studies have indicated that the level of lung radioactivity reflects the number of viable tumor cells in the lungs (for more information on this, see Ben-Eliyahu et al.  13).

Induction and Counting of Tumor Metastases 

Rats were lightly anesthetized with halothane, and 105MADB106 tumor cells were injected into their tail veins in 0.5 ml phosphate-buffered saline. Four weeks after tumor inoculation, rats were anesthetized, and lungs were removed and placed for 24 h in Bouin solution (72% saturated picric acid solution, 23% formaldehyde [37% solution], and 5% glacial acetic acid). After being washed in ethanol, visible surface metastases were counted by a researcher blinded to the donor of each lung. When the rats were killed, no signs of respiratory distress, body weight loss, or any noticeable illness-related behavior were observed. With the anticipated number of lung metastases, rats are not expected to show such signs before 8 weeks. 13 

Design and Procedure 

In a preliminary study, we developed the procedure for inducing hypothermia. Core temperature was monitored continuously until rats emerged from anesthesia. No immune measures were studied in this initial phase.

All further experiments were devoted to the study of immune measures related to NK activity. Rat temperatures were taken only three times: immediately after anesthesia and 1 and 2.5 h into the procedure. These measurements were conducted to verify adherence to the expected course of changes in core temperature.

In all experiments, rats were allocated to one of three groups: control, normothermia, and hypothermia. Rats from the control group remained awake and undisturbed in their home cages. Rats from the normothermia group were anesthetized and kept for 2.5 h at normal body temperature, as described before. Rats from the hypothermia group were anesthetized and their temperatures were reduced for 2.5 h, as described before. After this 2.5-h period, rats from the latter two groups were returned to their home cages. All immune- and tumor-related measures were taken from this time point (T0) and onward, and their timing is recorded in relation to T0.

In the first experiment, blood was drawn by cardiac puncture at 2 or 6 h to assess the activity and numbers of circulating NK cells (n = 54; 9–14 per group).

In the second experiment, rats were injected with radiolabeled MADB106 tumor cells at T0, or at 3 h. Lungs were removed 9 h after tumor injection so tumor retention could be assessed (n = 48; 9–15 per group).

In the third experiment, rats were injected intravenously with MADB106 cells at 2 h; 4 weeks later, lungs were removed and the number of metastases was counted (n = 57; 19 per group).

In the fourth experiment, each of the three experimental groups was subdivided and treated with saline or nadolol, a β-adrenergic antagonist. 38A first subcutaneous injection of the drug (0.4 mg/kg) or saline was given at the induction of anesthesia (−2.5 h), and a second identical injection was given together with radiolabeled MADB106 cells at 2 h. Lungs were removed so that tumor retention could be assessed 9 h after tumor injection (n = 73; 8–18 per group).

Statistical Analyses 

Natural killer cell activity was analyzed using repeated-measures analysis of variance for the different effector-to-target ratios and an α error of 0.05. In all other cases, analysis of variance was not used because markedly different variances characterized the different groups, thereby violating an analysis-of-variance assumption. Instead, we used unpaired Student t  tests to compare groups and the Bonferroni α-correction procedure to compensate for multiple comparisons. Thus, αcorrectedwas set at 0.015, allowing for as many as three independent comparisons ([1 −αcorrected]3> 95%). P  values indicate the significance levels of each t  test. Lytic units were calculated using the formula 100/ET30, where ET30is the effector-to-target ratio needed to lyse 30% of target cells. The regression exponential-fit method 39was used to infer ET30from the data.

Changes in Body Temperature 

In the hypothermia group, the average target core temperature of 30°C was achieved within 50 min and began to increase 40 min later, to 32°C by T0. Normal body temperature was regained gradually within 5 h of T0(fig. 1). The body temperature of normothermic animals remained at approximately 38°C before T0and typically decreased thereafter by as much as 1°C until rats had emerged from anesthesia completely. Hypothermic treatment was uniform in all subsequent experiments and resulted in very similar changes in body temperature.

Fig. 1. Body temperatures (mean ± SD) of anesthetized rats maintained at normothermia or hypothermia. Anesthesia was induced by intraperitoneal injection of 70 mg/kg sodium thiopental. The gray area denotes the 2.5-h period in which body temperature was controlled. 

Fig. 1. Body temperatures (mean ± SD) of anesthetized rats maintained at normothermia or hypothermia. Anesthesia was induced by intraperitoneal injection of 70 mg/kg sodium thiopental. The gray area denotes the 2.5-h period in which body temperature was controlled. 

Close modal

Activity and Numbers of Blood Natural Killer Cells 

At 2 h and 6 h, hypothermia significantly suppressed NKCA compared with levels in the control (P < 0.0001) and the time-matched normothermic (P < 0.05) groups (fig. 2A). Normothermia significantly suppressed NKCA at 6 h (P < 0.05), but not at 2 h (P = 0.068). Because NKCA at the two time points was similar in each group (fig. 2A), they were combined for further analysis using lytic units. Normothermia suppressed NKCA (expressed as the mean lytic units at 30% specific killing [LU30]± SEM) to 39% of control levels (from 49.30 ± 10.30 in the control group to 19.08 ± 3.03;P < 0.003). Hypothermia further suppressed NKCA to 15% of control levels (to 7.41 ± 3.24;P < 0.015)(fig. 2B).

Fig. 2. Animals were subjected to thiopental anesthesia in either normothermic or hypothermic conditions or served as controls. Blood was drawn 2 or 6 h after T0. (  A ) The percentage of specific lysis (mean ± SEM) at different effector-to-target ratios (natural killer:YAC-1 tumor cells). Natural killer cell activity in the two hypothermic groups was significantly less than in the control group and in the time-matched normothermic groups. (  B ) Natural killer cell activity expressed as lytic units at 30% specific killing (LU30; mean ± SEM). The two time points were joined because no differences were evident. *A significant difference (  P < 0.02). (  C ) The numbers of natural killer cells (NKR-P1bright) per microliter of blood (mean ± SEM). No significant differences were evident. 

Fig. 2. Animals were subjected to thiopental anesthesia in either normothermic or hypothermic conditions or served as controls. Blood was drawn 2 or 6 h after T0. (  A ) The percentage of specific lysis (mean ± SEM) at different effector-to-target ratios (natural killer:YAC-1 tumor cells). Natural killer cell activity in the two hypothermic groups was significantly less than in the control group and in the time-matched normothermic groups. (  B ) Natural killer cell activity expressed as lytic units at 30% specific killing (LU30; mean ± SEM). The two time points were joined because no differences were evident. *A significant difference (  P < 0.02). (  C ) The numbers of natural killer cells (NKR-P1bright) per microliter of blood (mean ± SEM). No significant differences were evident. 

Close modal

At the two time points when blood was drawn to assess NK activity, NK cell numbers were similar in all groups, ranging from 305 to 350 NK cells/μl (fig. 2C).

Lung Tumor Retention of MADB106 Cells 

After tumor injection at T0, normothermia increased the percentage of MADB106 tumor cells retained in the lungs from 0.40 ± 0.05% in the control group to 0.64 ± 0.09%(P = 0.024). This effect completely dissipated within 3 h (fig. 3A). Hypothermia, on the other hand, caused more than a twofold increase in lung tumor retention at T0(as much as 1.01 ± 0.23;P < 0.004) and at 3 h (as much as 0.94 ± 0.16%;P < 0.001). The increase caused by hypothermia compared with normothermia was also significant at T0and 3 h (P < 0.01 and P < 0.002, respectively).

Fig. 3. (  A ) The percentage of lung tumor retention (mean ± SEM). Animals were anesthetized and subjected to either normothermic or hypothermic conditions or served as controls. MADB106 cells were injected at T0or at 3 h. * A significant difference (  P < 0.02). (  B ) The number of lung tumor metastases (mean ± SEM) counted 4 weeks after MADB106 inoculation in normothermia, hypothermia, and in control animals. Tumor cells were injected at 2 h. * A significant difference (  P < 0.02). 

Fig. 3. (  A ) The percentage of lung tumor retention (mean ± SEM). Animals were anesthetized and subjected to either normothermic or hypothermic conditions or served as controls. MADB106 cells were injected at T0or at 3 h. * A significant difference (  P < 0.02). (  B ) The number of lung tumor metastases (mean ± SEM) counted 4 weeks after MADB106 inoculation in normothermia, hypothermia, and in control animals. Tumor cells were injected at 2 h. * A significant difference (  P < 0.02). 

Close modal

MADB106 Lung Metastases 

The average number of metastases identified in the hypothermia group was 4.68 ± 1.27, which is more than three times greater than in the control and normothermia groups, in which the metastases numbered 1.10 ± 0.4 and 1.47 ± 0.39, respectively (one-tailed test;P < 0.006 and P < 0.011, respectively)(fig. 3B).

Importantly, only 21% of the hypothermic animals were free of metastases, whereas in the normothermic and control groups, 36.8% and 52.6%, respectively, were free of metastases.

Effects of β-Adrenergic Blockade on Lung Tumor Retention 

Normothermia and hypothermia significantly increased lung tumor retention, from 0.53 ± 0.05% in the control group to 1.30 ± 0.16% and 1.57 ± 0.22%, respectively (fig. 4). Although the effect of hypothermia was larger than that of normothermia, this difference was not statistically significant in this experiment. Nadolol attenuated the effects of hypothermia, keeping tumor retention at 0.93 ± 0.07%(hypothermia saline vs.  hypothermia nadolol;P = 0.0055). This attenuation was not statistically significant in the normothermia condition (normothermia saline vs.  normothermia nadolol;P = 0.0329).

Fig. 4. The percentage of lung tumor retention (mean ± SEM) in animals inoculated with MADB106 cells. Rats were treated with nadolol or vehicle and assigned to a condition of either control, normothermia, or hypothermia. * A significant difference (  P < 0.02). 

Fig. 4. The percentage of lung tumor retention (mean ± SEM) in animals inoculated with MADB106 cells. Rats were treated with nadolol or vehicle and assigned to a condition of either control, normothermia, or hypothermia. * A significant difference (  P < 0.02). 

Close modal

To the best of our knowledge, this is the first study to systematically assess the effect of hypothermia under anesthesia on NK activity or tumor metastasis. Our findings show that 2.5 h of hypothermia during thiopental anesthesia markedly suppressed NK activity and promoted metastasis. Specifically, normothermic rats showed less suppression of NKCA than did hypothermic rats. This suppression occurred at the level of individual NK cell activity, because the numbers of NK cells per milliliter of blood were unaffected. Correspondingly, the retention of MADB106 tumor cells in the lungs and the number of lung metastases were significantly increased by hypothermia compared with rats in the control and the normothermia groups. The effects of normothermia (thiopental anesthesia alone) were smaller and reached significant levels only on one of the four occasions that the MADB106 was used. Finally, nadolol, a β-adrenergic antagonist, significantly attenuated the effect of hypothermia on lung tumor retention.

Overall, these findings suggest that the suppression of NKCA by hypothermia is robust and long-lasting and has consequences for the health of the animal. Specifically, NKCA was still markedly suppressed at 6 h, and lung tumor retention, reflecting cumulative clearance of tumor cells from 3 to 12 h, was greatly increased. This latter effect seems resistant to factors that influence tumor development at later stages, because lung metastases, counted 4 weeks after hypothermia, were four times as numerous in the rats in the hypothermia group compared with those in the control or normothermia groups. Considering the high sensitivity of the MADB106 to NKCA 12–14and our previous findings of parallel effects of various manipulations on NKCA and on resistance to MADB106 metastasis, 29,40the suppression of NKCA by hypothermia is likely to underlie the increase in MADB106 metastases.

The suppression of NKCA by hypothermia seems unrelated to the cooling of NK cells themselves or to the cooling of other leukocytes. The blood samples from all groups were kept at room temperature (23°C;i.e. , “hypothermia”) for approximately 3 h during the preparation for the assessment of NKCA. Natural killer cell cytotoxicity was then assessed in all samples at 37°C. Therefore, it appears that physiologic responses secondary to hypothermia are involved in the suppression of NKCA. Possible candidates are the various hormones and cytokines released after hypothermia, most prominently catecholamines, and to a lesser degree glucocorticoids and other hormones. 27,28 

It is our hypothesis that adrenergic discharge plays an important role in the suppression of both NKCA and resistance to MADB106 metastasis observed in the current study. Anesthesia and surgery have been found to increase epinephrine levels. Similarly, hypothermia, under surgical conditions and in nonanesthetized persons, has been reported to increase norepinephrine levels compared with normothermia. 27,28,In vitro  suppression of NKCA by catecholamines has been demonstrated by various groups, 41–43and we recently reported that in vivo β-adrenergic stimulation markedly suppresses NKCA per NK cell and consequently promotes tumor metastasis of the same tumor used in the current study. 29Correspondingly, β-adrenergic antagonists have been reported to block the suppression of both NKCA and tumor resistance caused by various manipulations. 44,45Finally, in the current study the β-adrenergic antagonist, nadolol, almost completely blocked the effects of thiopental anesthesia and hypothermia on the retention of the MADB106 tumor cells. Although several studies in humans have suggested that adrenergic stimulation can enhance NKCA, 34,46there are indications that this increase is attributable to a transient elevation in the number of circulating NK cells, and that the long-term effect of catecholamines on NK activity is suppressive. 33,34 

Suppression of NKCA per NK cell, when induced in vivo , is likely to dissipate with time during preparations for the in vitro  assessment of NKCA. 13When blood is withdrawn, serum is replaced with an artificial medium, and factors such as cytokines and hormones are removed. In vitro  studies using human blood have shown that the suppressive effects of catecholamines dissipate after hormone removal. 43Therefore, a prolonged delay before the assessment of NKCA might conceal the effects of hypothermia. In the current study, only 3 h elapsed between blood withdrawal and assessment of NKCA. The whole-blood assay, or other timesaving procedures, may be considered to obviate these obstacles in future clinical studies.

Although several surgical procedures in humans are characterized by levels of hypothermia similar to or deeper than those reached here, 47most are not. To clarify the clinical applicability of the current findings, a wide range of hypothermic levels and durations should be tested to identify the conditions in which hypothermia begins to suppress NKCA. Simultaneous assessment of neuroendocrine responses could suggest potential mediators. These important elaborations are beyond the scope of the current study and should be addressed in experimental and clinical conditions. An additional unresolved issue in this study is the possible effect of hypothermia on the pharmacokinetics of thiopental metabolism. Because this anesthetic had effects of its own, some of the effects of hypothermia could be attributed to elevated plasma concentrations of thiopental. Nevertheless, we estimate this factor to be of minor significance: Hypothermia prolonged the duration of anesthesia by only 15% but typically had a much greater effect than normothermia on the immune measures, some of which (e.g. , the number of metastases) were not affected at all by normothermia.

Recent studies have suggested the molecular mechanisms that make NK cells important for preventing metastasis. By the time malignant cells evolve to a metastasizing stage, they have usually suppressed the expression of functional major histocompatibility complex class I molecules, making them invulnerable to cytotoxic T cells. The mechanism by which NK cells recognize their targets, however, involves the detection of aberrations in, or the absence of, these molecules. 48This, and the fact that a large proportion of NK cells reside in the blood and the tissues targeted by metastasizing cells (e.g. , the lung and liver), make NK cells most important at this stage. 49,50Apart from lysing single tumor cells dispersed from the primary tumor, NK cells have been shown to infiltrate and diminish established micrometastases. 16 

Our findings suggest that hypothermia may contribute to the clinically well-documented observation that surgery suppresses NKCA. 1–4In the current study in rats, hypothermia and suppression of NKCA were also associated with decreased resistance to metastasis. Relations of this kind would be difficult to assess clinically, because the delay between tumor dissemination and detection of possible related metastases is long and variable. In contrast, suppression of immunity by surgery or hypothermia may lead to an outbreak of latent or opportunistic infections within days, a phenomenon that is indeed well documented clinically. 5,21In sum, it is as yet unclear to what extent the morbidity and immunosuppression observed after surgery is caused by hypothermia. Nevertheless, given the current findings, exposing patients to intraoperative hypothermia should be considered carefully when there is a high risk of viral infection or metastatic spread. Of particular clinical relevance is the removal of metastasizing tumors in procedures that tend to cause hypothermia, such as open abdominal surgeries. Because hypothermia is imperative in several lifesaving procedures, prophylactic measures, such as preoperative stimulation of the immune system or the perioperative use of β-blockers, should be considered.

Pollock RE, Lotzova E, Stanford SD: Surgical stress impairs natural killer cell programming of tumor for lysis in patients with sarcomas and other solid tumors. Cancer 1992; 70: 2192–202
Beilin B, Shavit Y, Hart J, Mordashov B, Cohn S, Notti I, Bessler H: Effects of anesthesia based on large versus small doses of fentanyl on natural killer cell cytotoxicity in the perioperative period. Anesth Analg 1996; 82: 492–7
Page GG, Ben-Eliyahu S, Liebeskind JC: The role of LGL/NK cells in surgery-induced promotion of metastasis and its attenuation by morphine. Brain Behav Immun 1994; 8: 241–50
Zoller M, Heumann U, Betzler M, Stimmel H, Matzku S: Depression of nonadaptive immunity after surgical stress: Influence on metastatic spread. Invasion Metastasis 1989; 9: 46–68
Holzheimer RG, Haupt W, Thiede A, Schwarzkopf A: The challenge of postoperative infections: Does the surgeon make a difference? Infect Control Hosp Epidemiol 1997; 18: 449–56
Nathens AB, Chu PT, Marshall JC: Nosocomial infection in the surgical intensive care unit. Infect Dis Clin North Am 1992; 6: 657–75
Pollock RE, Babcock GF, Romsdahl MM, Nishioka K: Surgical stress-mediated suppression of murine natural killer cell cytotoxicity. Cancer Res 1984; 44: 3888–91
Herberman RB, Ortaldo JR: Natural killer cells: Their roles in defenses against disease. Science 1981; 214: 24–30
Moretta L, Ciccone E, Poggi A, Mingari MC, Moretta A: Origin and functions of human natural killer cells. Int J Clin Lab Res 1994; 24: 181–6
Brittenden J, Heys SD, Ross J, Eremin O: Natural killer cells and cancer. Cancer 1996; 77: 1226–43
Lotzova E: Role of interleukin-2 activated MHC-nonrestricted lymphocytes in antileukemia activity and therapy. Leuk Lymphoma 1992; 7: 15–28
Ben-Eliyahu S, Yirmiya R, Liebeskind JC, Taylor AN, Gale RP: Stress increases metastatic spread of a mammary tumor in rats: Evidence for mediation by the immune system. Brain Behav Immun 1991; 5: 193–205
Ben-Eliyahu S, Page GG: In vivo assessment of natural killer cell activity in rats. Prog Neuro Endocrine Immunol 1992; 5: 199–214
Barlozzari T, Leonhardt J, Wiltrout RH, Herberman RB, Reynolds CW: Direct evidence for the role of LGL in the inhibition of experimental tumor metastases. J Immunol 1985; 134: 2783–9
Wiltrout RH, Herberman RB, Zhang S, Chirigos MA, Ortaldo JR, Green KM, Talmadge JE: Role of organ-associated NK cells in decreased formation of experimental metastases in lung and liver. J Immunol 1985; 134: 4267–74
Whiteside TL, Herberman RB: The role of natural killer cells in immune surveillance of cancer. Curr Opin Immunol 1995; 7: 704–10
Levy SM, Herberman RB, Maluish AM, Schlien B, Lippman M: Prognostic risk assessment in primary breast cancer by behavioral and immunological parameters. Health Psychol 1985; 4: 99–113
Schantz SP, Brown BW, Lira E, Taylor DL, Beddingfield N: Evidence for the role of natural immunity in the control of metastatic spread of head and neck cancer. Cancer Immunol Immunother 1987; 25: 141–8
Frank SM, Beattie C, Christopherson R, Norris EJ, Rock P, Parker S, Kimball AW Jr: Epidural versus general anesthesia, ambient operating room temperature, and patient age as predictors of inadvertent hypothermia. A NESTHESIOLOGY 1992; 77: 252–7
Forstot RM: The etiology and management of inadvertent perioperative hypothermia. J Clin Anesth 1995; 7: 657–74
Kurz A, Sessler DI, Lenhardt R: Perioperative normothermia to reduce the incidence of surgical-wound infection and shorten hospitalization. N Engl J Med 1996; 334: 1209–15
Wenisch C, Narzt E, Sessler DI, Parschalk B, Lenhardt R, Kurz A, Graninger W: Mild intraoperative hypothermia reduces production of reactive oxygen intermediates by polymorphonuclear leukocytes. Anesth Analg 1996; 82: 810–6
Biggar WD, Bohn DJ, Kent G, Barker C, Hamilton G: Neutrophil migration in vitro and in vivo during hypothermia. Infect Immunol 1984; 46: 857–9
Beilin B, Shavit Y, Rozumovsky J, Wolloch Y, Zaidel A, Bessler H: Effect of mild perioperative hypothermia on cellular immune response. A NESTHESIOLOGY 1998; 89: 1133–40
Sheffield CW, Sessler DI, Hunt TK: Mild hypothermia during isoflurane anesthesia decreases resistance to E. coli dermal infection in guinea pigs. Acta Anaesthesiol Scand 1994; 38: 201–5
Clardy CW, Edwards KM, Gay JC: Increased susceptibility to infection in hypothermic children: Possible role of acquired neutrophil dysfunction. Pediatr Infect Dis 1985; 4: 379–82
Frank SM, Higgins MS, Breslow MJ, Fleisher LA, Gorman RB, Sitzmann JV, Raff H, Beattie C: The catecholamine, cortisol, and hemodynamic responses to mild perioperative hypothermia. A randomized clinical trial. A NESTHESIOLOGY 1995; 82: 83–93
Frank SM, Higgins MS, Fleisher LA, Sitzmann JV, Raff H, Breslow MJ: Adrenergic, respiratory, and cardiovascular effects of core cooling in humans. Am J Physiol 1997; 272: R557–62
Shakhar G, Ben-Eliyahu S: In vivo beta-adrenergic stimulation suppresses natural killer activity and compromises resistance to tumor metastasis in rats. J Immunol 1998; 160: 3251–8
Katafuchi T, Take S, Hori T: Roles of sympathetic nervous system in the suppression of cytotoxicity of splenic natural killer cells in the rat. J Physiol 1993; 465: 343–57
Masera R, Gatti G, Sartori ML, Carignola R, Salvadori A, Magro E, Angeli A: Involvement of Ca2+-dependent pathways in the inhibition of human natural killer (NK) cell activity by cortisol. Immunopharmacology 1989; 18: 11–22
Dhabhar FS, Miller AH, McEwen BS, Spencer RL: Effects of stress on immune cell distribution. Dynamics and hormonal mechanisms. J Immunol 1995; 154: 5511–27
Kappel M, Tvede N, Galbo H, Haahr PM, Kjaer M, Linstow M, Klarlund K, Pedersen BK: Evidence that the effect of physical exercise on NK cell activity is mediated by epinephrine. J Appl Physiol 1991; 70: 2530–4
Schedlowski M, Falk A, Rohne A, Wagner TO, Jacobs R, Tewes U, Schmidt RE: Catecholamines induce alterations of distribution and activity of human natural killer (NK) cells. J Clin Immunol 1993; 13: 344–51
Chambers WH, Brumfield AM, Hanley-Yanez K, Lakomy R, Herberman RB, McCaslin DC, Olszowy MW, McCoy JP Jr: Functional heterogeneity between NKR-P1bright/Lycopersicon esculentum lectin (L.E.)bright and NKR-P1bright/L.E.dim subpopulations of rat natural killer cells. J Immunol 1992; 148: 3658–65
Ottenhof PC, Morales A, Baines MG: Quantitation of a whole blood assay for human natural killer cell activity. J Immunol Methods 1981; 42: 305–18
Ben-Eliyahu S, Page GG, Shakhar G, Taylor AN: Increased susceptibility to metastasis during pro-oestrus/oestrus in rats: Possible role of oestradiol and natural killer cells. Br J Cancer 1996; 74: 1900–7
Prichard BN, Tomlinson B: The additional properties of beta adrenoceptor blocking drugs. J Cardiovasc Pharmacol 1986; 8: 1–15
Pollock RE, Zimmerman SO, Fuchshuber P, Lotzová E: Lytic units reconsidered: Pitfalls in calculation and usage. J Clin Lab Anal 1990; 4: 274–82
Ben-Eliyahu S, Page GG, Yirmiya R, Shakhar G: Evidence that stress and surgical interventions promote tumor development by suppressing natural killer cell activity. Int J Cancer 1999; 80: 880–8
Whalen MM, Bankhurst AD: Effects of beta-adrenergic receptor activation, cholera toxin and forskolin on human natural killer cell function. Biochem J 1990; 272: 327–31
Takamoto TY, Hory Y, Koga H, Yokoyama MM: Norepinephrin inhibits human natural killer cell activity in vitro. Int J Neurosci 1991; 58: 127
Hellstrand K, Hermodsson S, Strannegard O: Evidence for a beta-adrenoceptor-mediated regulation of human natural killer cells. J Immunol 1985; 134: 4095–9
Irwin M: Stress-induced immune suppression. Role of the autonomic nervous system. Ann N Y Acad Sci 1993; 697: 203–18
Stefanski V, Ben-Eliyahu S: Social confrontation and tumor metastasis in rats: Defeat and beta-adrenergic mechanisms. Physiol Behav 1996; 60: 277–82
Tonnesen E, Tonnesen J, Christensen NJ: Augmentation of cytotoxicity by natural killer (NK) cells after adrenaline administration in man. Acta Pathol Microbiol Immunol Scand (C) 1984; 92: 81–3
Taggart DP, Fraser W, Gray CE, Beastall G, Shenkin A, Wheatley DJ: The effects of systemic intraoperative hypothermia on the acute-phase and endocrine response to cardiac surgery. Thorac Cardiovasc Surg 1992; 40: 74–8
Dohring C, Colonna M: Major histocompatibility complex (MHC) class I recognition by natural killer cells. Crit Rev Immunol 1997; 17: 285–99
Algarra I, Collado A, Garrido F: Altered MHC class I antigens in tumors. Int J Clin Lab Res 1997; 27: 95–102
Garrido F, Ruiz-Cabello F, Cabrera T, Perez-Villar JJ, Lopez-Botet M, Duggan-Keen M, Stern PL: Implications for immunosurveillance of altered HLA class I phenotypes in human tumours. Immunol Today 1997; 18: 89–95