Penn State College of Medicine, Milton S. Hershey Medical Center, 500 University Drive, H187, Hershey, PA 17033, USA
*Corresponding author. E-mail: acronin@psu.edu
Accepted for publication: July 28, 2003
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. After IRB approval and informed consent was obtained, 10 healthy volunteers underwent an 11 pm to 7 am infusion of saline, and at least 1 week later an infusion of 0.020.04 µg kg1 min1 remifentanil. Blood was collected at 7 am for measurement of NK cell cytotoxicity using a 51Cr release assay and measurement of NK cell number using fluorescent flow cytometry.
Results. Median and range of the total NK cell cytoxicity (KU ml1) was 745.0 (498.31483.6) on the control morning and 818.6 (238.51454.5) on the morning following the remifentanil infusion. Neither the number of NK cells ml1 (2.5x105 (1.4x1054.2x105) vs 2.7x105 (1.1x1054.4x105)) nor the cytotoxicity per 1000 NK cells (KU 1000 NK cells1) (3.0 (1.85.2) vs 2.9 (0.96.7)) changed between the control and remifentanil conditions.
Conclusions. An 8-h infusion of remifentanil did not affect NK cell activity in normal volunteers. This result differs from previous findings of morphine-induced NK cell activity suppression and fentanyl-induced NK cell activity enhancement in normal volunteers.
Br J Anaesth 2003; 91: 8059
Keywords: analgesics opioid, remifentanil; cells, natural killer cell activity
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In rats, equipotent systemically administered doses of morphine, fentanyl, and sufentanil cause similar reductions of NK cell function.2 Recently, remifentanil has also been shown to suppress NK cell function in the rat.5 In studies performed in healthy human volunteers, however, the effects of opioids on NK cell activity are inconsistent. Morphine suppresses NK cell activity in a dose-dependant manner.4 In contrast, fentanyl increases NK cell number and cytolytic function.6 7 The effect of remifentanil on NK cell activity has not been investigated in humans and is the subject of this study.
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Remifentanil titration
At least 2 days before the collection of data on NK cell number and function, subjects underwent titration of remifentanil to determine the dose that evoked a respiratory depressant effect (defined below) and to exclude subjects who experienced nausea or dysphoria. While subjects rested supine in a well-lighted room, an i.v. infusion of saline was initiated. Ventilatory frequency and arterial haemoglobin oxygen saturation (measured using pulse oximetry) were observed for 30 min. Mean of the values obtained at 20, 25, and 30 min were determined and these served as the baseline. A remifentanil infusion was then initiated at 0.02 µg kg1 min1 and increased to 0.04 µg kg1 min1 after 30 min. Ventilatory frequency and arterial haemoglobin oxygen saturation were monitored every 5 min for 30 min, and the means of the 20-, 25-, and 30-min values were determined. If the ventilatory frequency fell below 75% of baseline or the oxygen saturation fell below 97% of baseline, the infusion rate was reduced in 0.01 µg kg1 min1 increments until the ventilatory frequency and oxygen saturation were greater than or equal to these thresholds.
Experimental design
All subjects first underwent a control (saline) infusion and 13 weeks later a remifentanil infusion, with the subject blinded to the treatment order. For each infusion, the subject spent two consecutive nights in the General Clinical Research Center (GCRC). The first night was an acclimatization night. The purpose of this night was to ensure that subjects received normal sleep time on the night before data collection and to reduce the likelihood of psychological stress as a result of the unfamiliarity with the study environment. Subjects were admitted to the GCRC at 9 pm, an i.v. tubing was taped to their arm to simulate the i.v. that would be present the next night, and a pulse oximetry probe was placed on a finger. From 10 pm until 8 am subjects remained supine. Juice was offered at 10.30 pm. The lights, radio, and television were turned off at 11 pm. By 8.15 am the following morning, subjects were discharged and were asked to maintain their normal daytime activities and avoid alcohol.
Subjects returned to the GCRC at 9 pm for the subsequent saline or remifentanil treatment night. On this night, an i.v. catheter was placed in the proximal forearm and an infusion of saline was initiated at 80 ml h1. Exactly the same protocol described for the acclimatization night was followed on the treatment night, except that an i.v. catheter was placed and a saline or remifentanil infusion was initiated at the predetermined rate at 11 pm and continued until 7 am. Oxygen saturation was monitored throughout the infusions. At 7 am the infusion was discontinued and blood was drawn from the i.v. catheter for determination of NK cell function and number. All subjects were judged ready for discharge by 9 am.
Assay of NK cell cytolytic function
Six millilitres of blood was collected into a VacutainerTM CPT® tube (Becton-Dickinson, Franklin Lakes, NJ). Lymphocytes and monocytes were separated from the other cellular elements by centrifugation at 15001600 g for 20 min at 22°C, washed twice in RPMI 1640 media supplemented with 10% heat-inactivated foetal bovine serum, 100 µg ml1 streptomycin sulphate, 100 units ml1 penicillin, 50 µM 2-mercaptoethanol, 2 mM glutamine and resuspended to a volume of 6 ml in supplemented RPMI 1640 media. To determine NK cell lytic activity, 100 µl aliquots of these separated cells were placed in a 96-well V-bottom plate in serial 2-fold dilutions (1:1 to 1:128). Into each well, 2000 51Cr-labelled K562 cells were added. Into separate wells, an equal volume of 5% sodium dodecyl sulphate (SDS) detergent was added to produce the maximum possible K562 cell lysis (and release of 51Cr into the supernatant). Spontaneous (minimum) 51Cr release from 51Cr-labelled K562 cells in the absence of NK cell exposure was also determined by the addition of an equal volume of supplemented RPMI 1640 media alone. All wells were incubated for 21 h at 37°C in carbon dioxide 5%. After this incubation period, 50 µl of the supernatant from each well was added to 5 ml of ScintiVerse BD Cocktail (Fisher Scientific) and was counted for 1 min by liquid scintillation analysis to determine the counts per minute (cpm) in each sample. Each dilution of separated cells was assayed in triplicate, and the mean of the three values obtained were used in all further analyses. Cytotoxicity for each dilution of NK cells was expressed as the percentage of the specific lysis (achieved by SDS following removing of spontaneous 51Cr release):
With no compelling reason to choose any specific dilution as being more relevant than any other, we have chosen to include the data from all of the dilutions (1:1 to 1:128), and to describe the cytolytic activity of each blood sample as the total area under that curve, expressed as killing units (KU) (Fig. 1).
|
Statistical analysis
As the data were not normally distributed, Wilcoxon signed rank test was used to test the primary hypothesis that total NK cell cytolytic activity was decreased by remifentanil and to make additional comparisons of NK cell number and function between saline and remifentanil.
The sample size of 12 patients was based on previous studies that demonstrated a significant effect of morphine and of fentanyl on NK cell function in normal volunteers using seven to nine subjects per group.4 6 7 This sample size allowed for a 25% attrition rate to achieve a final subject number of at least 9.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Respiratory depression (ventilatory frequency <75% of baseline and oxygen saturation <97% of baseline) from remifentanil was present in three subjects at 0.02 µg kg1 min1, 2 subjects at 0.03 µg kg1 min1, and five subjects at 0.04 µg kg1 min1. All subjects demonstrated some degree of respiratory depression, but one subject did not meet the respiratory depression threshold (Table 2). Remifentanil caused a statistically significant decrease in ventilatory frequency (P=0.0001) and oxygen saturation (P=0.0017 two-tailed paired t-test).
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
It is possible that the effect of remifentanil on NK cells is different from that of the other mu opioid agonists, morphine and fentanyl. However, an alternative explanation for these varying results is differences in study design. A specific objective of the design of the current study was to control for factors such as psychological stress,8 sleep deprivation,9 phase of circadian rhythm,10 and amount of physical activity,11 all of which influence NK cell activity. The short half-life of remifentanil enables this study to evaluate the effect of several hours of a stable plasma opioid concentration on NK cells and permits titration of the opioid dose to a modest physiologic target effect. These conditions minimize possible confounding effects of increasing and decreasing opioid concentrations and dose-dependent side-effects such as nausea or dysphoria that might be associated with neuroendocrine effects on NK cells.
Unlike human studies, animal-based studies consistently demonstrate suppression of NK cell activity by opioids, including morphine, fentanyl, and remifentanil.2 5 12 NK cell suppression occurs following systemic or direct central nervous system administration of opioids to rats. Furthermore, microinjection studies have localized the site of action to the periaqueductal grey matter,13 and injection of subtype selective agonists has demonstrated that only the mu receptor mediates this opioid effect on NK cell activity.14 As morphine, fentanyl, and remifentanil are all mu agonists that cross the bloodbrain barrier, the animal studies do not help explain why these opioids have differing effects on NK cell activity in humans.
Morphine, fentanyl, and remifentanil have different plasma half-lives and metabolites that might help explain their differing effects on NK cell activity measured in vivo and ex vivo. This possibility is especially attractive because of the presence of opioid receptors on many lymphoid cell types.15 In this study, for example, continued metabolism of remifentanil by cholinesterases in the blood sample before NK cell analysis might have diminished the measured effect of remifentanil on NK cell activity (but not NK cell number).
However, consistent evidence from both animal and human studies cast doubt on this theory. In animals, systemic administration of N-methyl-morphine, an active morphine derivative that does not cross the bloodbrain barrier, does not suppress NK cell activity.16 In humans, the measured effect of systemically administered morphine or fentanyl on NK cells does not depend upon whether the assay is performed in opioid-containing or opioid-free conditions.4 Therefore, differences in plasma opioid metabolites or opioid concentrations in the blood sample during the NK cell assay are at best incomplete explanations for the reported differences in effect of morphine, fentanyl, and now remifentanil on NK cell activity in humans. Rather than a direct effect of opioids or their metabolites on NK cells in the plasma, the evidence from the animal and human studies points toward a neural-immune linkage to explain the effect of opioids on NK cells.
Because of the evidence indicating that the central nervous system is critical in connecting opioids and NK cells, this study aimed to rigorously control the level of opioid activity in the brain. We attempted to establish a uniform degree of central nervous system opioid activity by using the respiratory response as a functional assay. The minimum dose of remifentanil that achieved clear evidence of mu opioid receptor activation was chosen, and the pharmacokinetic properties of remifentanil enabled us to reliably maintain this stable plasma concentration for several hours. Therefore, the reported NK cell effects are not attributable to increasing or decreasing plasma opioid concentrations.17 These differences in study design might explain the differences in results among the studies of the effects of opioids on NK cell activity in human volunteers.
Another significant difference between this and previous studies of opioid effect on NK cell activity in humans is the choice of the control group. Previous studies4 7 compared the NK cell activity immediately before opioid administration (control) to that activity at time points during or following opioid administration, or used a separate group of five healthy volunteers as a control group.6 The current study is the first to use a placebo (saline) infusion and an opioid infusion in the same study subjects, thus allowing the use of paired comparisons at the same time of day following 8 h of inactivity in a recumbent posture.
Because the opioid effect on NK cells is most likely transmitted through a neuralimmune interaction, physical and psychological factors influencing subjective experiences are potentially significant. For this reason, effort was directed to acclimatize the subjects in our study to the study environment and to ensure a normal night of sleep before the treatment night. The lowest opioid dose that evoked a clear physiologic response was used in order to avoid common opioid side-effects such as nausea, with concomitant neuroendocrine alterations.18
Using rigorously controlled conditions, this study has found no evidence of a clinically significant effect of a low dose of remifentanil on NK cell activity in human volunteers. Whether remifentanil is different from morphine or fentanyl in this regard or whether the effects reported by others result from differences in study design needs further study. The clinical implication of these investigations remains unclear pending this information, and ultimately a study of opioid effects on NK cell activity in patients rather than normal volunteers.
![]() |
Conclusion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Acknowledgements |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Beilin B, Martin F, Shavit Y, Gale R, Liebeskind J. Suppression of natural killer cell activity by high-dose narcotic anesthesia in rats. Brain Behav Immun 1989; 3: 12937[Medline]
3 Carr D, Gebhart B, Paul D. Alpha adrenergic and mu-2 opioid receptors are involved in morphine-induced suppression of splenocyte natural killer activity. J Pharmacol Exp Ther 1993; 264: 117986[Abstract]
4 Yeager M, Colacchio T, Yu C, et al. Morphine inhibits spontaneous and cytokine-enhanced natural killer cell cytotoxicity in volunteers. Anesthesiology 1995; 83: 5008[ISI][Medline]
5 Sacerdote P, Gaspani L, Rossoni G, Panerai A, Bianchi M. Effect of the opioid remifentanil on cellular response in the rat. Int Immunopharmacol 2001; 1: 7139[CrossRef][ISI][Medline]
6 Jacobs R, Karst M, Scheinichen D, et al. Effects of fentanyl on immune functions in man. Int J Immunopharmacol 1999; 21: 44554[CrossRef][ISI][Medline]
7 Yeager M, Procopio M, DeLeo J, Arruda J, Hildebrandt L, Howell A. Intravenous fentanyl increases natural killer cell cytotoxicity and circulating CD(+)16 lymphocytes in humans. Anesth Analg 2002; 94: 949
8 Spiegel D, Sephton SE. Psychoneuroimmune and endocrine pathways in cancer: effects of stress and support. Semin Clin Neuropsychiatry 2001; 6: 25265[CrossRef][Medline]
9 Born J, Lange T, Hansen K, Molle M, Fehm HL. Effects of sleep and circadian rhythm on human circulating immune cells. J Immunol 1997; 158: 445464[Abstract]
10 Suzuki S, Toyabe S, Moroda T, et al. Circadian rhythm of leucocytes and lymphocytes subsets and its possible correlation with the function of the autonomic nervous system. Clin Exp Immunol 1997; 110: 5008[CrossRef][ISI][Medline]
11 Shinkai S, Konishi M, Shephard RJ. Aging, exercise, training, and the immune system. Exerc Immunol Rev 1997; 3: 6895[ISI][Medline]
12 Shavit Y, Lewis J, Terman G. Opioid peptides mediate the suppressive effect of stress on natural killer cell cytotoxicity. Science 1984; 223: 18890[ISI][Medline]
13 Weber R, Pert A. The periaqueductal gray matter mediates opiate-induced immunosuppression. Science 1989; 245: 18890[ISI][Medline]
14 Nelson C, Schneider G, Lysle D. Involvement of central mu- but not delta- or kappa-opioid receptors in immunomodulation. Brain Behav Immun 2000; 14: 17084[CrossRef][ISI][Medline]
15 Madden JJ, Whaley WL, Ketelsen D. Opiate binding sites in the cellular immune system: expression and regulation. J Neuroimmunol 1998; 83: 5762[CrossRef][ISI][Medline]
16 Shavit Y, Depaulis A, Martin F, et al. Involvement of brain opiate receptors in the immune-suppressive effect of morphine. Proc Natl Acad Sci USA 1986; 83: 71147[Abstract]
17 Kapila A, Glass PS, Jacobs JR, et al. Measured context-sensitive half-times of remifentanil and alfentanil. Anesthesiology 1995; 83: 96875[CrossRef][ISI][Medline]
18 Hogger P. Adverse effects of opioid analgesic treatment are correlated with a significant elevation in plasma epinephrine in healthy humans. Eur J Clin Pharmacol 2000; 56: 46368[CrossRef][ISI][Medline]