| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
 
|
|
|
||
|
|
© 2002 International Anesthesia Research Society
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Opioids, including fentanyl, are often administered to patients who may be at risk for the consequences of impaired immune function. We performed a clinical study to test the effects of the synthetic opioid fentanyl on human immune function. Participants received an IV fentanyl initial dose of 3 µg/kg followed by a 2-h IV infusion of 1.2 µg · kg-1 · h-1. Peripheral blood was drawn before and after fentanyl administration to test for neutrophil phagocytic function, neutrophil antibody-dependent cell cytotoxicity, natural killer cell cytotoxicity, percentage of lymphocyte populations, T-lymphocyte proliferative response, and in vivo antibody response to a pneumococcal vaccine inoculation given at the end of the fentanyl infusion. Fentanyl exposure under the conditions of this study caused a rapid and significant increase in natural killer cell cytotoxicity, which was coincident with an increase in the percentage of CD16+ and CD8+ cells in peripheral blood. Fentanyl did not significantly affect any of the other immune measurements.
IMPLICATIONS: Many previous studies have suggested that opioid drugs can impair immune resistance in patients who may be at risk for infection. This study suggests that the opioid fentanyl, when given to healthy humans without coexisting diseases, does not suppress immune resistance. On the basis of these results, the use of fentanyl should not be restricted because of concerns that it may suppress immune function.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Opioids are a mainstay of treatment for acute pain after surgery or injury and for many types of chronic pain, including cancer pain. Opioids are also a common substance of abuse, especially in the population of IV drug users (IVDUs) (1). In an environment of widespread and expanding opioid use, the immunologic effects of opioids have received considerable attention because of concerns that opioid-induced changes in immune function may affect the outcome of surgery or of a variety of disease processes, including bacterial or viral infections and cancer.
When tested in vitro, the immunologic effects of opioids depend on the particular opioid that is studied, on the immune cell population tested, and on the assay conditions (2,3). Opioids can produce a stimulated, depressed, or dose-dependent biphasic effect on immunity (2–4). This marked variability in observed responses is caused, in part, by the differing effects of endogenous compared with exogenous opioids, donor-specific responses, and the effects of opioid tolerance (4–6). Some in vivo studies in animals have suggested an immune-suppressive effect, particularly from chronic or large-dose opioid exposure (6,7). In vivo studies in humans are comparatively few. The population of IVDUs, which includes opiate users, may have depressed immune function, but the etiology of this finding is unclear because of the presence of multiple-drug use, high-risk behaviors, or coexisting infection (1,8). Acute exposure to morphine can temporarily suppress some aspects of immune function, notably natural killer cell cytotoxicity (NKCC) (9), but the clinical implications of this finding are unknown.
The synthetic opioid fentanyl is often prescribed for the management of acute pain of surgery and for the management of chronic malignant and nonmalignant pain. Fentanyl also has a greater potency than classic opioids such as morphine. For these reasons—the widespread use of fentanyl and its marked potency compared with classic opioids—we conducted a clinical investigation to evaluate the effect of IV fentanyl on tests of both innate and acquired immunity in humans and on the percentage of leukocyte subpopulations in peripheral blood.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study was approved by our IRB, and written, informed consent was obtained from all participants. Study participants (n = 7) were healthy human subjects, aged 22 to 45 yr, taking no medications known to alter immune function, and free of clinical infections for at least 3 mo. Participants were admitted to a hospital observation unit at 7:00 AM having had nothing to eat or drink since midnight. A peripheral IV infusion of lactated Ringer’s solution was started and maintained at 100 mL/h. Fentanyl was then administered as an IV initial dose of 3 µg/kg, followed by a continuous infusion at the rate of 1.2 µg · kg-1 · h-1 for 2 h. Participants were kept under observation for 2 h after fentanyl administration and then discharged home by following a standard discharge protocol for surgical outpatients.
Neutrophil Antibody-Dependent Cell Cytotoxicity
As previously described (10), polymorphonuclear neutrophils (neutrophils) were isolated by density gradient centrifugation from heparinized whole blood. The neutrophils were then washed, red blood cells were lysed with sterile water, and the remaining neutrophils were resuspended in serum-free medium. Chicken red blood cells (cRBCs) were labeled with 51Cr, washed twice, and adjusted to 0.5 x 106 cells per milliliter. The following were then added in triplicate to 96-well plates to achieve appropriate experimental conditions: 50 µL of rabbit anti-cRBC antibody (final concentrations, 0.047 to 1.5 µg/mL), 50 µL of neutrophils, and 50 µL of cRBCs. The plate was centrifuged and incubated at 37°C in 5% CO2 for 4 h. The percentage specific cytotoxicity was then calculated as follows:
|
|
where cpm (experimental) = counts after incubation of effector cells with target cells, cpm (maximum) = counts after detergent lysis of target cells, and cpm (control) = counts after incubation of target cells in medium alone.
Neutrophil Phagocytosis
In duplicate, 1 x 106 neutrophils in 100 µL, 5 x 106 Candida (kept in continuous culture), 50 µL of autologous serum, and a balance of serum-free medium to a final volume of 500 µL were added to 1.5-mL polypropylene tubes. The suspensions were incubated with constant agitation for 15 min in a 37°C water bath and placed on ice. A wet mount slide was made and read for percentage phagocytosis of 200 neutrophils per sample (number of neutrophils with ingested yeast/total number of neutrophils x 100 = percentage phagocytosis). Results are reported as the average of the duplicates.
NKCC
As previously described (9), peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation, washed, and resuspended in medium. Aliquots of K562 target cells were labeled with 51Cr, washed, and resuspended to a final concentration of 1 x 105 cells per milliliter. A total of 1 x 104 K562 cells and effector cells to achieve the desired effector/target (E/T) ratios were added in 100-µL aliquots to 96-well plates and measured in triplicate. After incubation at 37°C with 5% CO2 for 4 h, 70 µL of supernatant was removed from each well and counted for 1 min on a gamma counter to determine the percentage specific cytotoxicity by using the same calculation reported for neutrophil antibody-dependent cell cytotoxicity (see previously). Specific cytotoxicity results represent the average of the three determinations.
T-Cell Proliferative Response
PBMCs containing primarily lymphocytes with <5% monocyte contamination were isolated as above and frozen in liquid nitrogen until assayed. Cells were thawed and resuspended, and by use of trypan blue dye exclusion, the cell count was adjusted to a concentration of 1 x 106 viable cells per milliliter. Cells were stimulated with concanavalin A in microtiter assay plates at final concentrations of 1 and 5 µg/mL of concanavalin A, incubated in 5% CO2 at 37°C for 3 days, and tested in triplicate (with positive and negative control wells). On the second day, cells were pulsed with 1 µCi per well of tritiated thymidine. Cells were harvested on the third day on filter paper and placed in a scintillation vial for radioactivity determination on a beta counter. Results are reported as mean cpm with correction of the mean by subtracting the mean cpm of background wells.
Antibody Response to Antigen Inoculation
One week before the study date, serum was drawn by peripheral venipuncture and frozen at -70°C for subsequent analysis. At the end of the fentanyl administration, subjects received an IM injection of a polyvalent pneumococcal vaccine (Pneumovax®; Merck & Co., Inc., West Point, PA). A second serum sample was drawn 21 days after pneumococcal vaccine administration and stored at -70°C for subsequent analysis. Paired serum samples were forwarded for commercial enzyme-linked immunosorbent assay analysis of antipneumococcal antibody responses (Specialty Laboratories, Santa Monica, CA).
Leukocyte Subset Analysis
PBMCs were isolated as above. Murine monoclonal antibodies (mAbs) to cell surface antigens were obtained as purified antibody from the Dartmouth Medical School Hybridoma Facility. Specific mAbs included OKT4 to CD4, OKT8 to CD8, AML-2-23 to CD14, 3G8 to CD16, mAb 32 to CD64, mAb W6/32 to human leukocyte antigen class I, HNK-1 to CD57, OKT3 to CD3, and FMC63 to CD19. Binding of each mAb was detected with fluorescein isothiocyanate-conjugated, affinity-purified goat antimouse immunoglobulin (Ig)G and IgM F(ab')2 fragments. Approximately 0.5 x 106 cells were incubated for 30 min at 4°C with saturating amounts of mAb. Incubation with primary antibodies was performed in the presence of 4.8 x 10-5 M human IgG to saturate type I high-affinity Fc receptors (FcRI) for IgG and to block nonspecific binding of mAbs. After the incubation with primary antibody, cells were washed and incubated with fluorescein isothiocyanate goat antimouse IgG antibody for an additional 30 min at 4°C. Cells were again washed, fixed in paraformaldehyde, and stored at 4°C in the dark until analyzed in duplicate for expression of each surface molecule. Flow cytometry was performed on a flow cytometer (FACScan; Becton Dickinson, Franklin Lakes, NJ) by using logarithmic amplification of the green fluorescence scale. Ten thousand cells from each antibody-stained sample were analyzed, and the data from duplicate samples were averaged.
A power analysis for study size was based on results from previous work in which small clinical concentrations of morphine showed significant decrements in NKCC after in vivo exposure (9). With these data, we determined that we would have 90% power to detect a 35% decrement in NKCC with a 5% level of significance in a treatment group in which six participants completed the study. Paired intragroup comparisons were performed with a Student’s t-test with, each participant serving as his or her own control. Intragroup multiple pairwise comparisons were analyzed by analysis of variance. A P value <0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Seven participants (four men and three women) were studied. The mean age and weight of participants were 36.6 yr and 71 kg, respectively. Other than sedation, nausea, or vomiting, which occurred in all seven subjects, no significant side effect of fentanyl was observed.
Neutrophil studies were performed at baseline, at the end of the fentanyl infusion, and at 1 and 24 h after the fentanyl infusion. As reported for morphine (10), a consistent decrement in neutrophil antibody-dependent cell cytotoxicity activity was noted 24 h after fentanyl exposure. This was most apparent at the midrange of antibody concentrations but was not statistically significant at any antibody concentration (Fig. 1). Neutrophil phagocytosis was unchanged after fentanyl exposure (33% [5.5%], 33% [5.5%], 30% [5.8%], and 28% [4.3%] phagocytosis [± SE] at each measurement time).
|
T-lymphocyte proliferative responses were measured at baseline and at 1 and 24 h after the fentanyl infusion. The proliferative response was not affected by fentanyl exposure, with proliferative responses virtually unchanged from baseline at 1 and 24 h after fentanyl administration (Fig. 2).
|
NKCC was tested at baseline, at the end of the fentanyl infusion, and at 1 and 24 h after the fentanyl infusion. Fentanyl produced a significant increase in NKCC at the end of the infusion—an effect that was significant at all E/T ratios (Fig. 3A). By 1 h after the fentanyl infusion, NKCC was not significantly different from baseline at any E/T ratio. Because the marked change in natural killer activity after fentanyl was unexpected, a control comparison group (n = 6) was subsequently studied. These subjects received only an IV infusion of lactated Ringer’s solution under the same conditions as those used for fentanyl administration. Measurements of NKCC were made in the control subjects at baseline, at the end of the lactated Ringer’s infusion, and 1 h after the infusion. An infusion of lactated Ringer’s solution alone had no effect on NKCC (Fig. 3B).
|
Antibody responses to four different pneumococcal serotypes (serotypes 3, 7F, 9N, and 14) were tested by comparing baseline values (4911 [± 1151] ng/mL, 1378 [± 556] ng/mL, 1318 [± 633] ng/mL, and 2723 [± 985] ng/mL [± SE], respectively) in serum with values measured 21 days after the IM pneumococcal vaccine inoculation values (12,251 [± 2812] ng/mL, 2950 [± 649] ng/mL, 2333 [± 888] ng/mL, and 5071 [± 2510] ng/mL [± SE], respectively). Baseline and postinoculation values for these antibody titers are similar to those reported in the literature (11) and are not significantly different from results in a group of healthy individuals inoculated under the same conditions (12).
Leukocyte subset analysis was performed at baseline, at the end of the fentanyl infusion, and at 24 h after the fentanyl infusion. Fentanyl exposure was associated with a significant increase in the percentage of CD16+ lymphocytes in peripheral blood at the end of the fentanyl infusion (Table 1). The number of CD16+ lymphocytes had decreased to 21% by 24 h later, a value that was not significantly different from the baseline value (P = 0.08). Fentanyl exposure was also associated with a significant increase in the percentage of CD8+ lymphocytes measured at the end of infusion, a change that persisted at 24 h (Table 1).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study demonstrated that short-term in vivo exposure to fentanyl increases NKCC directed against a tumor cell target. This effect seemed to be caused by an increase in the number, but not the activity, of natural killer cells (NKCs) in the peripheral blood. All other measures of immunity were not affected. The results have important implications for several reasons. First, the study investigated clinically-relevant doses of fentanyl compared with much smaller doses previously measured under similar conditions (13). Second, the study was conducted in healthy participants to avoid the confounding effects of coexisting diseases, surgical trauma, or pain. Third, fentanyl can be a substance of abuse in the population of IVDUs who do not experience trauma or chronic pain (similar to the participants in this study), but who are at risk for immune dysfunction and infection (1). Finally, the study used several commonly tested measures of both innate and acquired immune function to allow for comparison with results in other published reports.
Previous publications have reported that opioids can affect neutrophil function. Tubaro et al. (7) reported that morphine inhibited neutrophil phagocytosis, both in mice and also in rabbits, and that these effects were accompanied by increased susceptibility to experimental infection. Subsequent reports showed that IVDUs maintained on morphine had diminished neutrophil phagocytic activity over a period of time relative to controls or to a population of IVDUs maintained on methadone (14). Our results do not entirely support these findings, and this may be because different opiate dosages were used for animal studies and because the immune effects of varying opioid dosages do not always correlate with behavioral effects that test for analgesia in animals (15). Other studies attest to a variable dosage effect of opioids on neutrophil function. Sharp et al. (16) showed that at smaller doses, opioid peptides could stimulate neutrophil oxidative burst, but that this effect had an inverse dose-response relationship with decreasing stimulation at larger doses. Another important difference is the use of opsonizing serum or antibody-directed killing in both of our neutrophil assays. Sowa et al. (17) reported that extremely small concentrations of morphine could inhibit phagocytic ingestion of nonopsonized cryptococcal organisms by swine microglia in vitro. Opsonization, as used for our phagocytic assay, may alter the observed effects of an opioid on phagocytic function and raises the question of whether opioids may differentially affect granulocyte elimination of pathogenic organisms at different sites in the body outside the blood. Our study was limited to testing neutrophils in peripheral blood, where they may demonstrate functional responses different from those seen in other body compartments (3). Finally, we did not study the immune effects of chronic opioid exposure.
Fentanyl exposure produced a rapid and significant increase in NKCC that was not caused by the IV infusion alone, because a series of control treatments showed no effect on NKCC. This result differs from a previously reported study in which a 24-hour IV infusion of morphine caused a temporary, but significant, reduction in peripheral blood NKCC (9). Because the previously reported morphine-induced depression of NKCC was first observed two hours after the onset of a morphine infusion and because the total duration of the fentanyl infusion in this study was also two hours, the increase in NK activity seen with fentanyl may be a unique pharmacodynamic effect distinct from that of morphine and not a consequence of the more rapid onset of fentanyl effects. Although the mechanism of increased NKCC is uncertain, we did observe a significant increase in peripheral blood CD16+ cells, which are predominantly NKCs, suggesting that recruitment of NKCs into the central circulation may explain the enhanced NK activity. Prior reports have documented an acute increase in NKCs after an infusion of catecholamines in healthy subjects in what seemed to be a ß-adrenergic-mediated recruitment of CD16+ cells from peripheral sites into the central circulation (18). In a report similar to ours, Jacobs et al. (13) studied seven subjects, who received a single IV bolus dose of fentanyl (0.2 µg/kg) that was much smaller than the dose that we used. They found that fentanyl significantly increased circulating CD16+/CD56+ cells (NKCs) within 15 minutes. This effect remained significant 30 minutes after the fentanyl bolus (later time points were not reported) and was reversed by naloxone, thereby suggesting involvement of opioid receptors. They also reported that fentanyl did not affect NKCC, but they used in vitro incubation of effector cells with fentanyl, whereas our study used in vivo exposure of effector cells to fentanyl followed by ex vivo testing.
T-lymphocytic proliferation is a commonly measured test of T-cell responsiveness that is opioid sensitive. This effect has been most extensively documented in animals, where alterations in lymphocyte numbers and function have been repeatedly shown after opioid exposure, especially morphine exposure (3,6,19). This direct action of opioids on peripheral immune cells is specific to different body compartments (3) and is not part of an analgesic response (20). As an in vitro assay, the lymphocyte proliferative response bypasses several steps for in vivo antigen processing and does not test B-cell responsiveness and antibody production. To fully test the integrity of the in vivo antigen-antibody response, we tested the response to an IM inoculation with a polyvalent polysaccharide vaccine (pneumococcus). Although there was no specific comparison group, the levels of antibodies at baseline and after inoculation were similar to those reported in other studies of healthy humans (11,12) and suggest no effect of acute fentanyl exposure on in vivo antigen responsiveness.
Overall, this study failed to demonstrate significant effects of fentanyl on human immune activity, with the single exception of a rapid increase in peripheral blood natural killer activity that may have been caused by recruitment of NKCs into the central circulation from noncirculating sites. The important limitations of this investigation include the requirement for ex vivo tests of peripheral blood samples, the short duration of opioid exposure, and, by design, the absence of an interaction with pain or disease. From a clinical standpoint, the administration of opioids per se may be a less important consideration than the clinical setting in which they are administered and their interactions with other components of the human inflammatory response.
|
Acknowledgments |
|---|
Supported in part by National Institutes of Health Grant DA-09162 (MPY).
|
Footnotes |
|---|
Presented in part at the annual meeting of the American Society of Anesthesiologists, Orlando, FL, October, 1998.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- Rouveix B. Opiates and immune function: consequences on infectious diseases with special reference to AIDS. Therapie 1992; 47: 503–12.[Web of Science][Medline]
- Williamson SA, Knight RA, Lightman SL, Hobbs JR. Differential effects of ß-endorphin fragments on human natural killing. Brain Behav Immun 1987; 1: 329–35.[Medline]
-
Lysle DT, Coussons ME, Watts VJ, et al. Morphine-induced alterations of immune status: dose dependency, compartment specificity and antagonism by naltrexone. J Pharmacol Exp Ther 1992; 265: 1071–8.
[Abstract/Free Full Text] - Millar DB, Hough CJ, Mazorow DL, Gootenberg JE. ß-Endorphin’s modulation of lymphocyte proliferation is dose, donor, and time dependent. Brain Behav Immun 1990; 4: 232–42.[Web of Science][Medline]
- Roy S, Loh HH. Effects of opioids on the immune system. Neurochem Res 1996; 21: 1375–86.[Web of Science][Medline]
-
Molitor TW, Morilla A, Risdahl JM, et al. Chronic morphine administration impairs cell-mediated immune responses in swine. J Pharmacol Exp Ther 1991; 260: 581–6.
[Abstract/Free Full Text] - Tubaro E, Borelli G, Croce C, et al. Effect of morphine on resistance to infection. J Infect Dis 1983; 148: 656–66.[Web of Science][Medline]
- Novick DM, Ochshorn M, Kreek MJ. In vivo and in vitro studies of opiates and cellular immunity in narcotic addicts. In: Friedman H, Specter S, Klein TW, eds. Drugs of abuse, immunity, and immunodeficiency. New York: Plenum Press, 1991: 159–70.
- Yeager MP, Colacchio TA, Yu CT, et al. Morphine inhibits spontaneous and cytokine-enhanced natural killer cell cytotoxicity in volunteers. Anesthesiology 1995; 83: 500–8.[Web of Science][Medline]
- Yeager MP, Yu CT, Campbell AS, et al. Effect of morphine and ß-endorphin on human Fc receptor-dependent and natural killer cell functions. Clin Immunol Immunopathol 1992; 62: 336–43.[Web of Science][Medline]
-
Jackson LA, Benson P, Sneller VP, et al. Safety of revaccination with pneumococcal polysaccharide vaccine. JAMA 1999; 281: 243–8.
[Abstract/Free Full Text] -
Procopio MA, DeLeo JA, Rassias AJ, et al. In vivo effects of general and epidural anesthesia on human immune function. Anesth Analg 2001; 93: 460–5.
[Abstract/Free Full Text] - Jacobs R, Karst M, Scheinichen D, et al. Effects of fentanyl on cellular immune functions in man. Int J Immunopharmacol 1999; 21: 445–54.[Web of Science][Medline]
- Tubaro E, Avico U, Santiangeli C, et al. Morphine and methadone impact on human phagocytic physiology. Int J Immunopharmacol 1985; 7: 865–74.[Web of Science][Medline]
- Nelson CJ, Dykstra LA, Lysle DT. Comparison of the time course of morphine’s analgesic and immunologic effects. Anesth Analg 1997; 85: 620–6.[Abstract]
-
Sharp BM, Keane WF, Suh HJ, et al. Opioid peptides rapidly stimulate superoxide production by human polymorphonuclear leukocytes and macrophages. Endocrinology 1985; 117: 793–5.
[Abstract/Free Full Text] - Sowa G, Gekker G, Lipovsky MM, et al. Inhibition of swine microglial cell phagocytosis of Cryptococcus neoformans by femtomolar concentrations of morphine. Biochem Pharmacol 1997; 53: 823–8.[Web of Science][Medline]
- Schedlowski M, Falk A, Rohne A, et al. Catecholamines induce alterations of distribution and activity of human natural killer (NK) cells. J Clin Immunol 1993; 13: 344–51.[Web of Science][Medline]
-
Pruett SB, Han YC, Fuchs BA. Morphine suppresses primary humoral immune responses by a predominantly indirect mechanism. J Pharmacol Exp Ther 1992; 262: 923–8.
[Abstract/Free Full Text] - Hamra JG, Yaksh TL. Equianalgesic doses of subcutaneous but not intrathecal morphine alter phenotypic expression of cell surface markers and mitogen-induced proliferation in rat lymphocytes. Anesthesiology 1996; 85: 355–65.[Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  I. Tegeder and G. Geisslinger Opioids As Modulators of Cell Death and Survival--Unraveling Mechanisms and Revealing New Indications Pharmacol. Rev., September 1, 2004; 56(3): 351 - 369. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Volk, M. Schenk, K. Voigt, S. Tohtz, M. Putzier, and W. J. Kox Postoperative Epidural Anesthesia Preserves Lymphocyte, but Not Monocyte, Immune Function After Major Spine Surgery Anesth. Analg., April 1, 2004; 98(4): 1086 - 1092. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Cronin, N. M. Aucutt-Walter, T. Budinetz, C. P. Bonafide, N. A. DiVittore, V. Gordin, H. G. Schuler, and R. H. Bonneau Low-dose remifentanil infusion does not impair natural killer cell function in healthy volunteers Br. J. Anaesth., December 1, 2003; 91(6): 805 - 809. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. N. Raja, J. A. Haythornthwaite, M. Pappagallo, M. R. Clark, T. G. Travison, S. Sabeen, R. M. Royall, and M. B. Max Opioids versus antidepressants in postherpetic neuralgia: A randomized, placebo-controlled trial Neurology, October 8, 2002; 59(7): 1015 - 1021. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
