Opioids and opiates: analgesia with cardiovascular, haemodynamic and immune implications in critical illness


Patricia E. Molina MD, PhD, LSUHSC Physiology, 1901 Perdido St., New Orleans, LA 70112, USA.
(fax: (504) 568 6158; e-mail: pmolin@lsuhsc.edu).


Traumatic injury, surgical interventions and sepsis are amongst some of the clinical conditions that result in marked activation of neuroendocrine and opiate responses aimed at restoring haemodynamic and metabolic homeostasis. The central activation of the neuroendocrine and opiate systems, known collectively as the stress response, is elicited by diverse physical stressor conditions, including ischaemia, glucopenia and inflammation. The role of the hypothalamic–pituitary–adrenal axis and sympathetic nervous system in counterregulation of haemodynamic and metabolic alterations has been studied extensively. However, that of the endogenous opiates/opioid system is still unclear. In addition to activation of the opiate receptor through the endogenous release of opioids, pharmacotherapy with opiate receptor agonists is frequently used for sedation and analgesia of injured, septic and critically ill patients. How this affects the haemodynamic, cardiovascular, metabolic and immune responses is poorly understood. The variety of opiate receptor types, their specificity and ubiquitous location both in the central nervous system and in the periphery adds additional complicating factors to the clear understanding of their contribution to the stress response to the various physical perturbations. This review aims at discussing scientific evidence gathered from preclinical studies on the role of endogenous opioids as well as those administered as pharmacological agents on the host cardiovascular, neuroendocrine, metabolic and immune response mechanisms critical for survival from injury in perspective with clinical observations that provide parallel assessment of relevant outcome measures. When possible, the clinical relevance and corresponding scenarios where this evidence can be integrated into our understanding of the clinical implications of opiate effects will be examined. Overall, the scientific basis to enhance clinical judgment and expectations when using opioid sedation and analgesia in the management of the injured, septic or postsurgical patient will be discussed.


The re-establishment of homeostasis after injury involves activation of host defence mechanisms for self-protection against toxic inflammatory processes and tissue repair that are under tight neuroendocrine regulation. Evidence suggests that disruption of these tightly regulated and balanced pro-inflammatory host responses contributes to the morbidity and mortality associated with the post-injury period. These neuroendocrine mechanisms responsible for the control of haemodynamic, metabolic and immune counterregulatory responses to injury are sensitive to the use of anaesthetics and analgesics. Resuscitation interventions during the immediate post-injury period are centred on bleeding control and fluid replacement aimed at restoration of tissue perfusion pressure to preserve tissue metabolic function, immune competence and regenerative capacity. Pain alleviation, frequently based on opiate agonists, is central to the medical management following traumatic injury, a surgical intervention as well as in acutely ill individuals. Although analgesic and sedative pharmacological treatments are frequently used in the management of these patients, little is known regarding their impact on the immediate and subsequent counterregulatory responses to injury despite their potential for interacting with the neuroendocrine control of host defence mechanisms involved in restoring homeostasis.

Identification of the impact of the use of these therapeutic approaches in a clinical setting is complicated by multiple factors including the multiple pharmacologic therapies, environmental conditions leading to the disruption of normal circadian rhythms, underlying health status of the individual, age, gender and lifestyle prior to the time of injury. Thus a vast amount of the information available is derived from preclinical studies from which inferences and extrapolations to the clinical setting must be made. The role of endogenous and pharmacologic opiate receptor activation has been dissected using animal models that resemble a clinical presentation in which both the nervous system as well as the immune system are challenged accordingly and in which the outcome from neuroimmune interaction affects the host response. Preclinical studies have provided a wealth of information on the specific host defence, haemodynamic and metabolic responses that are affected by opioid receptor activation resulting either from their pharmacological administration as well as from their endogenous release during trauma, infection or haemorrhagic shock. The evidence gathered from those preclinical studies strongly suggests that opioids exert modulatory effects on several of the critical mechanisms involved in restoring homeostasis.

Neuroendocrine response to injury

Acute insults to the integrity of the host such as acute blood loss, infection or surgical interventions require adaptation involving the synchronized interaction of multiple neuronal and endocrine pathways geared at restoring homeostasis and ensuring survival. The integrated haemodynamic, metabolic, behavioural and immune responses that allow adaptation of the host are referred to as the stress response [1]. Hypotension and hypovolaemia resulting from severe blood loss as well as tissue injury elicit the immediate activation of the autonomic nervous system and the hypothalamic–pituitary–adrenal (HPA) axis. Central to the counterregulatory response triggered by blood loss is the increased sympathetic nerve activity [2] resulting in catecholamine release from the adrenals and noradrenaline turnover in peripheral organs where they exert paracrine and hormonal stimulation. The immediate activation of the sympathetic nervous system (SNS) and the resulting release in catecholamines both from the adrenal medulla as well as from nerve terminals at target organs play central roles in increasing cardiac contractility, heart rate and peripheral vasoconstriction [3]. In addition to mediating rapid mobilization of energy substrates from the liver, muscle and adipose tissue they exert well-recognized effects on metabolic and immune responses as well. In addition, blood loss stimulates effector neurones in the hypothalamus, increasing secretion of corticotrophin-releasing hormone (CRH) [4, 5], oxytocin [6] and arginine vasopressin (AVP) [7–9]. The enhanced activation of these descending neuroendocrine pathways results in increased circulating levels of noradrenaline, adrenaline, AVP, adrenocorticotrophic hormone (ACTH), glucocorticoids and β-endorphin [10–12].

Activation of endogenous opioid pathways in response to injury

Opiate receptor activation occurs as a natural response to injury resulting from their increased release in response to injury, infection, trauma and surgery (Fig. 1). In addition to the post-translational processing of pro-opiomelanocortin (POMC) in the anterior pituitary, their tissue localized release affects sensory, immune and metabolic responses during traumatic injury. Opioid peptides, the natural ligands at opioid receptors belong to three families of peptides derived from distinct genes. These peptide families and their precursor proteins POMC, pro-enkephalin, or pro-dynorphin have been well characterized in the central nervous system (CNS) and neuroendocrine system. Differential processing yields the opioids β-endorphin, enkephalins and dynorphins, which in turn exhibit different affinity and selectivity for the three opioid G-protein-coupled receptors: μ (β-endorphins and enkephalins), δ (enkephalins and endorphins) and κ (dynorphin) [13] coupled to the inhibitory G-protein (Gi/o) [14]. Ligand binding results in inhibition of adenylyl cyclase, decreasing intracellular cyclic adenosine monophosphate levels, and protein kinase A activity.

Figure 1.

Activation of the opioid system with stress. The neuroendocrine response to stress involves activation of two core systems; the hypothalamic–pituitary–adrenal (HPA) axis and the sympathetic nervous system involved in orchestrating counterregulatory responses. Activation of the HPA axis by corticotrophin-releasing factor results in post-translational cleavage of pro-opiomelanocortin (POMC) to adrenocorticotrophin, α-MSH (melanocyte stimulating hormone) and β-endorphin and their subsequent release into the systemic circulation. β-Endorphin exerts effects in the central nervous system as well as in peripheral tissues including cells of the immune system. In addition, peripheral tissues including the adrenal glands produce endogenous opioids that are also released into the systemic circulation or may affect neighbouring cells through paracrine mechanisms. Furthermore, POMC-related and pro-enkephalin-derived opioid peptides and the enzymes necessary for their processing in similar fashion to that of the pituitary have been identified in immune cells (Fig. 2). The activation of the sympathetic nervous system, shown in the figure only stimulating adrenal medulla release of catecholamines also produces tissue-localized co-release of noradrenaline and opioid peptides (not shown in the figure) at target organs. Whilst the central purpose of this stress response activation is to restore cardiovascular and haemodynamic stability, mobilize stored energy sources to sustain the increased metabolic demands of the host and ensure that immune competence and tissue repair can proceed, endogenous opioids may affect several of these either directly or indirectly. Whilst the impact of haemodynamic and cardiovascular responses as well as on immune responses is not beneficial to the host, it appears to confer stress-induced analgesia during the period of acute injury.

Preclinical studies show that circulating levels of β-endorphin, an index of endogenous opioid system activation, are significantly elevated following muscle crush injury, fixed-pressure haemorrhagic shock and lipopolysaccharide (LPS) administration in animal models [15]. Similar stress-induced elevations in the production and release of endogenous opioids have been demonstrated in clinical studies. Patients with coronary artery disease have augmented myocardial and circulating concentrations of β-endorphin following myocardial ischaemia and reperfusion during coronary angioplasty [16]. Surgical stress has also been shown to be associated with increased endorphin release. Patients undergoing elective coronary artery bypass grafting surgery with cardiopulmonary bypass under general anaesthesia show increased β-endorphin release [17]. Similarly, oral surgery [18] as well as elective gynaecological and abdominal surgery [19, 20] have been shown to result in increased β-endorphin levels in both male and female patients. In addition, elevated levels of endogenous opioids have been demonstrated in intubated critically ill patients receiving artificial respiration [21]. Interestingly, however, stress-induced activation of endogenous opioid release appears to be absent in these patients, as endobronchial lavage or suctioning did not produce any changes in levels of β-endorphin, suggesting an altered neuroendocrine regulatory status. Greater increases in circulating levels of β-endorphin have also been reported in during the postoperative period in patients with septic shock and sepsis than those without infection, but also in the critical care unit [22]. Similar to the dysregulation of neuroendocrine function observed in ventilated patients, severity of sepsis was associated with altered patterns of opioid, cortisol and ACTH responses. Together, these observations indicate that the upregulation of the endogenous opioid pathway in response to injury, stress and infection observed in preclinical studies is an accurate reflection of the changes that occur in the clinical setting. Furthermore, they render validity to the use of animal models to better understand the contribution of their endogenous release as well as that of exogenous pharmacological stimulation is appropriate to enhance our understanding of their overall role.

Contribution of endogenous opioids to overall neuroendocrine response

The balanced activation of opioid pathways in response to injury, integrated with the parasympathetic and SNSs and HPA axis activation results in dynamic regulation of host defence mechanisms vital for immune competence and tissue repair. Together, these neuropeptides mediate the haemodynamic and metabolic responses involved in restoring homeostasis. Whilst the activation of the SNS is easily recognized as critical in the control of the cardiovascular and haemodynamic counterregulatory response, the relevance of endogenous opioid activation and release has been less evident.

Endogenous opioid activation has been traditionally considered to be associated with anti-nociception [23] and in some cases with an overall bio-behavioural feeling of well-being [24]. Various preclinical rodent and large animal models of stress including forced cold water swim, electrical foot-shock, exposure of a pup to an adult rodent, immobilization, hypoglycaemia and fixed-pressure haemorrhagic shock have demonstrated that increased release of endogenous opioids produces a phenomena known as stress-induced analgesia [25–28]. This analgesic effect is largely mediated through the μ-opioid receptor and to a lesser extent by δ-opioid receptors [29, 30]. Similar role of endogenous opioids in stress-induced analgesia has been demonstrated in humans [31]. Stress induced by anticipation of pain produced by noxious foot shock has been shown to progressively increase the threshold for nociception in healthy volunteers [32]. Furthermore, heterotopic painful thermal stimuli produces a naloxone-sensitive decrease in the sensation of pain elicited by electrical stimulation of the sural nerve at the ankle of healthy volunteers [33]. The release of endogenous opiates has been shown to be closely associated with the analgesic effects. A negative correlation was observed between α-endorphin levels and individual pain scores in patients undergoing extracorporeal gallstone lithotripsy [34]. Furthermore, whilst β-endorphin levels rose in patients who perceived pain as well, the rise was only significant in those patients who did not report pain. The role of endogenous opioids in mediating analgesic effects has also been demonstrated for acupuncture [35]. Electro-acupuncture increases release of enkephalin, β-endorphin and dynorphin [36] resulting in effective therapy for several kinds of chronic pain including low back pain and diabetic neuropathic pain [37]. Thus, one could speculate that endogenous opiate activation and the resulting analgesic effects they mediated would be a ‘beneficial’ or ‘desired’ component of the stress response. However, additional opiate-mediated cellular responses would not appear to be so clearly beneficial to the compromised host (Fig. 2).

Figure 2.

Opiate-mediated immunomodulation. Opiates can exert their effects on immune function either directly on cells of the immune system or indirectly by affecting the release or effectiveness of other modulators of immune function. In addition to the central opioid production and release, cells of the immune system are also capable of processing and releasing pro-opiomelanocortin (POMC) and pro-enkephalin (PENK)-derived peptides. Their release is stimulated by corticotrophin-releasing hormone (CRH), viruses, lipopolysaccharide (LPS) and catecholamines. The locally produced opioid peptides can bind to opiate receptors on cells of the immune system (macrophages, lymphocytes and mast cells) where they exert immunomodulatory effects including reduced expression and release of adhesion molecules, tumour necrosis factor (TNF), substance P and calcitonin gene-related peptide (CGRP). Opiates also produce immunosuppressive effects on T cells including inhibition of induction of delayed-type hypersensitivity reactions and cytotoxic T-cell activity, modulation of T-cell antigen expression, and depression of responses to T-cell mitogens. Locally produced opioid peptides can also bind to afferent sensory neurones producing peripheral anaesthesia and affecting neurotransmitter (NT) release. Similar immunomodulatory and analgesic effects are achieved with pharmacologic administration of opiate receptor agonist such as morphine and fentanyl. MC, mast cells; ICAM; intercellular adhesion molecule; ROS, reactive oxygen species; β-EN, β-endorphin; Met-EK, met-enkephalin.

Opiate modulation of haemodynamic and cardiovascular responses

Whilst the pressor, inotropic and chronotropic effects elicited by activation of the SNS exert beneficial responses during a period of acute blood loss and tissue injury, the results from several studies suggest that the activation of the opioid system may counteract some of these effector responses during the recovery phase to acute blood loss and injury. Amongst the most relevant endogenous opioid-mediated effects on the response to traumatic injury are those that impact on haemodynamic regulation during blood loss. Studies in rodent models of haemorrhagic shock have demonstrated that opiate receptor antagonists ameliorate haemodynamic instability [38, 39]. Those findings have been confirmed by those from our laboratory. Our studies have shown that opioid receptor antagonist administration to conscious rodents did not alter prevailing mean arterial blood pressure levels but increased the volume of blood loss necessary to produce profound hypotension [40]. Others have shown that treatment with opiate antagonists improves arterial blood pressure and cardiac function in several animal models. Anaesthetized cynomologus monkeys bled to an average blood pressure of 45 mmHg showed increased mean arterial blood pressure and left ventricular change in pressure over time (LV, dP/dtmax), an index of cardiac function when treated with opiate antagonists [41]. In cats, naloxone treatment maintained mean arterial blood pressure at a higher value compared with vehicle-treated animals during the resuscitation phase [42]. In anaesthetized pigs, haemorrhaged to 50% of basal mean arterial blood pressure, naloxone was shown to increase peripheral vascular resistance thereby reversing shock-induced hypotension [43]. Furthermore, central opiate receptor activation has been demonstrated to decrease the volume of blood loss required to decrease blood pressure in anaesthetized rats subjected to stepwise haemorrhage [44]. Similar protective effects are elicited by central administration of β-endorphin antiserum to haemorrhaged rats and α-methyldopa-induced hypotension [45]. The role of endogenous opioids in modulation of haemodynamic responses has been further supported by reports of correlation of β-endorphin levels with impaired cardiac output [46, 45]. Although the role of endogenous opiates in outcome from shock and trauma has been addressed by several clinical studies, there are several limitations to the overall interpretation of the findings including sample size, heterogeneity of the population studied and lack of consistent identification of meaningful outcome measures. However, a considerably extensive meta-analysis review of the literature indicates that opiate antagonist treatment does improve mean arterial blood pressure in shock patients [47].

The mechanism involved in mediating the salutary effects of opiate antagonists has not been completely elucidated. Several possibilities can be identified based on both preclinical controlled experiments as well as by clinical observations, including direct and indirect mechanisms [48]. The ubiquitous presence of opiate receptors both in the CNS and in peripheral tissues [49], as well as the central and peripheral production [50] and release of these mediators, facilitates both an endocrine and paracrine contribution to modulating vascular responses. These opiate-mediated effects could be attributed to both central as well as peripheral interactions with the release of counterregulatory hormones as well through an interaction with target organ responsiveness.

Amongst the mechanisms that could be involved in mediating the haemodynamic salutary effects observed following the administration of opiate antagonists is enhanced catecholamine release [51] or effectiveness of their pressor effects [52]. This has been attributed to the presence of presynaptic κ, δ and μ-opioid receptors, with receptor-specific inhibitory effects on neurotransmitter release, specifically with release of noradrenaline being predominantly under μ-opiate receptor modulation [53]. Locally released endogenous opioid peptides can decrease noradrenaline release in a naloxone-sensitive manner [54], thereby centrally regulating excitability of the SNS. However, not all of the salutary effects of opiate antagonists can be explained by modulation of circulating catecholamines levels.

An additional mechanism through which opioids can alter cardiovascular function is through their interaction with adrenergic receptor function, particularly in the heart. Recent studies have demonstrate that opioid-mediated cardiovascular effects are the result of central opioid-mediated alteration in sympathetic control of cardiac function [55] as well as from direct opiate receptor-mediated responses in cardiac myocytes through endocrine and paracrine mechanisms. Opioid peptides are co-released from noradrenergic nerve terminals in the heart [56]. In addition, a localized endogenous opioid system has now been well characterized and described in mammalian cardiac tissue. Pro-enkephalin synthesis has been demonstrated in the isolated rat working-heart preparation [57]. Pro-dynorphin expression has been demonstrated in adult cultured rat ventricular cardiac myocytes and it has been shown to be upregulated in response to the potassium chloride challenge [58]. The expression of the final products (enkephalin and dynorphin) as well as the presence of δ- and κ-opiate receptors in these cells has also been demonstrated. Recent studies suggest that opiate receptor stimulation inhibits cardiac excitation–contraction coupling, but may also render the heart protection from hypoxic and ischaemic injury. These opioid effects on cardiac function have been partly attributed to interaction and ‘cross-talk’ between opiate and β-adrenergic receptors in cardiac myocytes where they are co-expressed [59].

Alternative mechanisms include a prevention of vagal activation during haemorrhagic shock, improving myocardial function and compensation from the hypotensive insult [60–62]. This would be in accordance with the observations of opioid-mediated increase in activation of vagal signalling in rodent models [63, 64]. Increased vagal activation during acute blood loss would result counterproductive in the process of restoring mean arterial blood pressure and tissue perfusion. The implications of these observations in the clinical setting, and in particular for clinical care are less well defined. Retrospective studies indicate that adverse cardiovascular effects consisting of inappropriate heart rate responses to hypotension caused by morphine sulphate are rare [65]. Indirect evidence indicates that opiate antagonists play a role in removing vagal inhibitory tone. Studies in opioid-dependent patients demonstrated that naltrexone treatment during the detoxification process increased heart rate and mean arterial blood pressure [66]. Recently heart rate variability has been identified as a useful tool for investigating the effect of drugs on the autonomic nervous system in the clinical setting. Using this approach, studies have shown that morphine administration prior to surgery is associated with parasympathetic dominance [67]. Patients scheduled for elective surgery under general anaesthesia showed decreased low frequency to high frequency (LF/HF) ratio, indicating a shift towards parasympathetic dominance. Similar observations have been reported in surgical patients undergoing elective surgical procedures when opiate sedation was used [68–70]. Together, these reports indicate that opiates favour a parasympathetic over sympathetic tone dominance strongly supporting this as an important mechanism for the cardiovascular depressor effects of opiate agonists and indirectly providing an explanatory mechanism for the beneficial effects of opiate antagonists.

Preclinical and clinical evidence suggests that endogenous opioid release, though exerting anti-nociceptive effects may negatively affect several of the neuroendocrine mechanisms involved in the host haemodynamic and cardiovascular counterregulatory response to shock and trauma. Because the regulatory pathways involved in mediating these cardiovascular responses to injury and blood loss also modulate host defence mechanisms the extent of opiate-mediated alterations in host response goes beyond that of cardiovascular responses.

Inflammatory response to injury

The period immediately following acute injury is characterized by an immediate upregulation of circulating and tissue pro-inflammatory gene expression (particularly in the lungs and spleen) [71, 72] which is associated with enhanced myeloperoxidase activity reflecting increased neutrophil influx, production of reactive oxygen species and apoptotic cell death all conducive to tissue injury [73–75]. Several underlying mechanisms are thought to be responsible for these tissue responses, including tissue destruction, hypoperfusion and endothelial activation [76–78]. This early post-traumatic inflammatory response is essential for tissue repair and to establish immune competence. However, when uncontrolled either in magnitude or duration, it is followed by a phase characterized by alterations of both cellular and humoral immunity [79–81] and inhibition of stimulated pro-inflammatory cytokine release [82, 83]. Several lines of evidence indicate that the early pro-inflammatory cytokine upregulation contributes to the development of this syndrome either by synergistic actions or by priming or predisposing the host to subsequent injury. Thus the pro/anti-inflammatory cytokine balance, which should mediate tissue repair and recovery, if uncontrolled, can produce tissue injury on one spectrum and immunosuppression on the other extreme [84–86]. A generalized immunosuppressed state results in an increased host susceptibility to infection [87, 88], delayed restoration of homeostasis and increased tissue injury leading to multiple organ failure [89, 90]. The course and progression of the host recovery from traumatic injury and the integrity of its response to a secondary challenge is directly related to the effective control of the immediate pro-inflammatory responses to the initial insult.

The course of an initial traumatic injury followed by impaired host response to systemic challenge reflecting the clinical scenario in which trauma victims unable to respond to life-threatening infections or sepsis during the post-injury period frequently develop multiple organ failure [91, 92] has been modelled in preclinical studies [93]. Using this model, we have dissected the role of the sympathetic and opioid pathways on the response to trauma-haemorrhage. Our results indicate that the regulated initiation and termination of tissue pro-inflammatory response is under neuroendocrine control through direct neurotransmitter release at the target organ as well as indirectly through endocrine factors including catecholamines, glucocorticoids, neuropeptides and opioid peptides. Thus, the same neuroendocrine effectors that control cardiovascular and haemodynamic responses to injury are also implicated in modulation of pro-inflammatory responses and as such can determine progression to tissue injury or repair [94, 95].

Neuroendocrine opioid modulation of host defence responses

The contribution of the SNS to the control of the pro-inflammatory responses to injury has been investigated by several laboratories including our own. The results from these studies have provided evidence that enhanced catecholamine release, limits the magnitude of the early pro-inflammatory response to haemorrhagic shock [96]. The modulation of tissue cytokine responses does not appear to be exclusively under noradrenergic modulation. Similar to what appears to be a close interaction between the opioid and noradrenergic systems in control of cardiovascular function, our studies suggest a close interaction between the noradrenergic and endogenous opioid pathways in modulating the magnitude of the early pro-inflammatory responses to haemorrhagic shock. We have seen that the removal of central nervous system noradrenergic tone results in attenuated endogenous opioid activation as well as suppression of the tissue pro-inflammatory response to haemorrhagic shock in conscious unrestrained rats. These findings suggest that central opioid activation requires central noradrenergic activation. In addition, they indicate that enhanced opioid activation during haemorrhagic shock contributes to the localized pro-inflammatory response. This potential role for endogenous opiates was further supported by results from subsequent rodent studies where we have demonstrated that opiate receptor inhibition with naltrexone blunted the haemorrhage-induced upregulation of tissue cytokine response. Thus, in addition to affecting cardiovascular responses to trauma-haemorrhage, endogenous opioid pathway activation also modulates the tissue pro-inflammatory responses. Hence, in this trauma-haemorrhage model, enhanced sympathetic activity exerts a suppressive effect on the haemorrhage-induced increase in tissue cytokine expression, whilst opiate system activation counteracts this noradrenergic suppression of pro-inflammatory response to haemorrhagic shock. Based on these observations, one can conclude that opiate-based therapeutic interventions may alter the host innate counterregulatory response and negatively affect recovery from traumatic injury.

The effects of opiates and opioid peptides on various aspects of host defence has been reported for some time [97–99]. Opiate-mediated immune effects have been postulated to result from either direct interaction with opioid receptors on cells of the immune system [100, 101] or indirectly, through the activation of opioid receptors within the central nervous system and the resulting modulation of HPA axis and the SNS [102]. Reports from in vivo and in vitro studies indicate that opiate receptor stimulation exerts suppression of multiple components of the host immune defence response including natural killer (NK) cell activity [103], neutrophil complement and immunoglobulin receptor expression [104], chemokine-induced chemotaxis [105] and phagocytosis [106]. Confirmation of the findings from these preclinical and in vitro studies in the clinical setting can be made by integrating the observations from human studies under different scenarios, particularly in patients receiving opiate analgesia, a widespread approach in the management of critically ill patients [107, 108]. Although the most frequently associated side-effects of opiate analgesia are nausea, vomiting, and pruritus, several lines of evidence obtained from in vivo animal studies and cell lines suggest that pharmacological opiate treatment affects immune function [109–111]. In parallel to the prolonged suppression in phagocytic capacity of circulating neutrophils isolated from morphine-treated animals observed in our animal studies, clinical studies have demonstrated a more prolonged suppression in phytohaemagglutinin (PHA)-induced lymphoproliferation during the postoperative period in morphine-treated patients undergoing abdominal surgery for uterine carcinoma [112]. Additional alterations in immune function observed with opioid-based anaesthesia are shifts in circulating lymphocyte phenotype. Studies in patients undergoing minor elective orthopaedic surgery showed significant enhancement of tumour necrosis factor-α and interleukin-1β responsiveness to LPS and PHA stimulation in whole blood cultures after induction of anaesthesia followed by a decrease in circulating lymphocytes, a significant increase in the percentage of CD4+ T lymphocytes and a significant decrease of NK cells all of which occurred prior to the surgical intervention [113]. In other studies, patients undergoing abdominal hysterectomy who received epidural sufentanil showed decreased NK cell activity at 1 and 3 days following the surgical procedure [114]. Moreover, studies in healthy volunteers administered with analgesic doses of intravenous morphine for 36 h showed significant suppression of NK cell cytotoxicity which persisted for 24 h after termination of morphine infusion [115].

However, not all clinical studies have obtained results supporting a deleterious effect of opiate analgesia on immune function. Some studies have reported only modest and transient alterations in NK cell cytotoxicity and lymphocyte phenotype. Short-term fentanyl administration to healthy volunteers has been shown to produce a rapid and significant increase in NK cell cytotoxicity coincident with an increase in the percentage of CD16+ and CD8+ cells in peripheral blood without significantly affecting any of the other immune measurements [116]. In contrast, patient-controlled intravenous opioid analgesia treatment to patients with extensive spine surgery produced a decrease in circulating CD4/CD8 lymphocyte ratio during the postoperative period [117]. These observations suggest differential opiate-mediated modulation of immune responses by depending not only on the health status of the host but also on the route and duration of administration.

It is difficult to predict if the alterations in immune function identified in healthy patients undergoing elective surgical procedures is conducive to immunopathology. Furthermore, interpretation of these observations is further riddled with differences in patient populations, duration, route and dose of anaesthetic administration as well as conditions under which immune function was evaluated. The results from our preclinical studies show that morphine treatment results in a significant impairment in the recovery of neutrophil phagocytic capacity following muscle crush injury followed by fixed-pressure haemorrhage in rats. This was associated with leucocytosis with a shift in distribution towards increased neutrophil counts. Based on these results, one can speculate that even modest alterations from normal function of lymphocytes can potentially affect host response mechanisms to a subsequent infectious challenge during the postoperative or recovery period. Furthermore, whilst function may appear to be preserved or only moderately affected by the administration of opioids, the stress, inflammatory and metabolic response to the injury produced by surgery, trauma or infection may unmask the underlying mechanisms that are affected by the presence of opioids in this patient population. Though this has not been adequately explored, some differences in the route of opioid administration have been observed in association with infection incidence, with a greater risk of wound infection in patients receiving epidural opioid analgesia than in those receiving intravenous opioid delivery [118]. A careful analysis of the impact of opiate analgesia on the incidence of infectious complications following traumatic injury, surgery or in critically ill patients warrants further study.

Opioid modulation of neuroendocrine responses

Opiate modulation of cardiovascular, haemodynamic and immune responses in the injured, postoperated or septic patient is not limited to one of direct opiate-receptor effects at the target organ (i.e. heart, vasculature, immune cells). An important mechanism through which these responses could be modulated is through opiate interaction with the neuroendocrine response to the stress of infection, blood loss or tissue injury, particularly, as a result of the use of opiate-based analgesia and sedation in the critically ill patient. Opiate sedation in the critical care setting is used to increase comfort and reduce the patient's stress response [119]. This pharmacological suppression of the stress response of the injured or critically ill patient is frequently desirable for the treating physician and can potentially be beneficial and protective from anxiety, dysphoria and ischaemic events [120]. Although this is common practice, the limited information on the impact of drug selection on outcome is well recognized [121]. Recent evidence would strongly caution against the indiscriminate suppression or attenuation of the central neuroendocrine axis involved in mediating this stress response. Evidence indicates that the responsiveness of core components of the stress response such as activation of the HPA axis and adrenal responsiveness are necessary as part of the defence mechanism of the individual [122]. Clinical studies show that the plasma cortisol response to CRH stimulation is impaired in nonsurvivors with severe sepsis, compared with those patients who survive [123]. Because the HPA response is already compromised in critically ill patients, one would expect that further addition of opioid-mediated analgesia and sedation would only contribute to further disruption in this system [124]. Recent understanding of the dynamic alteration in endocrine function in critically ill patients [125] would further caution against therapy geared at suppressing a stress response, particularly when several endocrine systems have been identified to reach near exhaustion during severe illness [126]. Because of the relevance of the neuroendocrine system in modulating cardiovascular, metabolic and immune host responses to injury, infection, surgical interventions and other insults, one can speculate that any disruption in the normal response, activation or termination of neuroendocrine pathways will contribute negatively impact outcome (Fig. 3).

Figure 3.

Opiate-mediated effects. Endogenous opioids and pharmacologically administered opiate analogues bind to Gi/o-protein-coupled receptors leading to the inhibition of adenylyl cyclase (AC), reduced levels of protein kinase A (PKA) activity and decreased phosphorylation of target proteins. In addition, opiate receptor activation by ligand binding leads to an increase in the conductance of an inwardly rectifying K+ channel and a decrease in Ca influx through the L-type Ca channel. Opiates exert their effects in multiple central and peripheral target cells producing respiratory depression, analgesia, negative inotropic effect, immunosuppression, etc. cAMP, cyclic adenosine monophosphate; C, catalytic subunit; RII, regulatory subunit.

Opioid modulation of metabolic responses

The immune responses to injury, infection and surgical stress and the process of healing and tissue repair are energy-requiring processes requiring active mobilization of energy substrate stores [127]. This metabolic response is characterized by hyperglycaemia, secondary to decreased glucose clearance and increased hepatic glucose production, increased proteolysis, particularly of the gut and enhanced lipolysis and has been examined in detail in mongrel dogs [128]. Whilst the enhanced substrate turnover plays a vital role in sustaining the enhanced energy needs during the recovery and healing period, its uncontrolled activation is also conducive to unwanted secondary effects during the post-injury period including insulin resistance [129] and muscle wasting [130]. Previous rodent studies from our laboratory have provided evidence suggesting that opiate receptor activation contributes to this systemic catabolic response to injury [131]. Of particular interest is the hyperglycaemic response elicited by opiate administration both in animal models [132] as well as in diabetic and nondiabetic patients [133] which can make the management of critically ill patients more complex. Recently, unidirectional cross talk between opiate and insulin signalling whereby opiate agonists activate signalling cascades that lead to an insulin-resistant state has been proposed to contribute to the hyperglycaemic effects of opiates based on observations on transfected cell lines [134]. The hyperglycaemic effect elicited by opiate administration is likely to further complicate glycaemic control in the intensive care unit. This would be an undesired effect based on the association between hyperglycaemia and adverse outcomes of critically ill patients following traumatic injury or surgical procedures [135], particularly based on the observations that metabolic control achieved by tight blood glucose control with insulin, leads to a dramatic decrease in morbidity and mortality [136]. Thus, the potential metabolic impact of opiate-based pharmacotherapy in critically ill patients needs to also be closely examined, as it could preclude appropriate glucose control in this setting.

Pain and opioids: from traumatic injury to the critically ill patient

Despite all the preclinical and clinical observations, the need for analgesia in the control of pain and discomfort is essential in the management of injured victims. Currently, selection of analgesic therapy is based on anti-nociceptive effectiveness, yet little is known about their potential interactions with the host counterregulatory mechanisms involved in restoring homeostasis. In particular, opiate analgesia a frequently integral component of the management of the traumatic injury victim leads to effective pain control, yet possible undesired immune effects remain to be studied systematically. However, the cardio-respiratory opiate-mediated adverse side-effects have been well recognized [137] and strategies such as the use of ketamine in combination with morphine have been proposed to reduce the doses of morphine required to produce adequate pain control [138]. Ketamine, an N-methyl-d-aspartate receptor antagonist, with potent anaesthetic effects produces a state of ‘dissociative anaesthesia’, maintains the respiratory drive and supports the systemic arterial blood pressure. Because of these characteristics, its combination with morphine has been proposed as an approach to decrease opiate use during the postoperative period. Nevertheless, clinical studies have provided conflicting reports on whether some benefit is obtained. Whilst some have provided evidence that ketamine reduces cumulative morphine consumption during patient-controlled analgesia following laparotomy [139] and provides rapid and sustained improvement in morphine analgesia without deleterious side-effects in patients undergoing elective surgical procedures [140], others have shown no major benefits after major abdominal surgery [141, 142]. The overall benefit of this approach remains subject of debate [143–145]. In addition, whether this drug combination would provide immunoprotection in the injured or critically ill patient is unclear. The results from our in vivo rodent studies as well as those from in vitro studies from other investigators indicate that, ketamine affects several aspects of the immune response [146–149]. More recently, a new generation of opioid drugs is emerging. This class of drugs selectively activates opioid receptors outside the central nervous system, thus avoiding all centrally mediated, unwanted effects. Interestingly though the central depressive effects may be obviated, the peripheral immune-related response is still likely to be conserved [150]. In this respect, peripherally active kappa opiate-receptor agonists exhibiting potent analgesic and anti-inflammatory effects have been developed [151]. Whilst in terms of the potential benefits in conditions of chronic subacute inflammation these may still not be indicated in critically ill or septic patients.

Few studies have investigated the impact of opiate analgesia on the outcome from traumatic injury. Preclinical studies show that morphine results in dose-dependent enhanced mortality from fixed-pressure haemorrhage in rodents [152]. Studies from our laboratory have demonstrated that morphine administration at doses that produce effective analgesia results in significant impairment in haemodynamic counterregulatory responses to haemorrhagic shock resulting in prolonged hypotension and increased morbidity and mortality. This was manifested by a long-lasting impact on the ability of the host to respond to a subsequent inflammatory challenge. Furthermore, those studies also showed that opiate analgesia markedly increased the resuscitation fluid requirements needed to restore mean arterial blood pressure to levels comparable with those of control animals. In this respect, increasing fluid resuscitation volumes abrogate a significant component of the deleterious effects of opiate analgesia. The relevance of this finding is the fact that use of predetermined volumes of fluid resuscitation may not be adequate for morphine-treated patients, requiring a more accurate assessment of mean arterial and perfusion pressures during the resuscitation period. This is of importance given that in the clinical setting analgesia is frequently initiated early on in the process of assessment of trauma victims, potentially altering the specific resuscitation fluid requirements.

The observations from both clinical and preclinical studies all point towards an overall immunosuppressive effect of opiate pharmacotherapy which are all conducive to increased susceptibility to bacterial and viral infections [109]. Nevertheless, there are cases as with the exaggerated inflammatory response in patients who develop SIRS where moderate inhibition of the pro-inflammatory response could be of potential benefit. One could speculate that opiate sedation could provide a ‘silencing’ approach to a dysregulated response. However, caution and knowledge derived from previous attempts at suppressing the inflammatory response which have failed to provide reliable beneficial results [153] should preclude this approach. Several pitfalls of the use of biological response modifiers in critically ill patients have been identified. Most importantly, the physiological response in the critically ill, stressed patient is complex with dynamic interaction of multiple redundant and balancing systems. The heterogeneity of the patient population with regard to the life-style factors, comorbid diseases, genetic polymorphisms, age, gender and overall psychological state at the time of injury determines the course, progression and outcome with specific therapy. All these factors in a constantly changing pathophysiological state increase the difficulty of identifying beneficial effects of therapy as well as time frame during which they can be effective.

Summary and conclusions

Our understanding of the impact of analgesic agents on haemodynamic and inflammatory counterregulatory responses to haemorrhage, trauma and infection is incomplete and strongly based on preclinical studies without direct confirmation in the clinical setting. The results from those studies indicate that opiate receptor activation resulting from either the endogenous release of opioid peptides or pharmacological administration of opiate-based analgesics for pain management exert modulatory effects on neuroendocrine, haemodynamic, metabolic responses involved in restoring homeostasis following acute injury, haemorrhagic shock and infection [40, 154]. Whilst some of the opiate effects observed in animal studies appear to be of greater severity than those identified in the clinical setting, the different mechanisms involved in the pathogenesis of shocked, traumatized or septic patients as well as the differential magnitude of the specific immune, cardiovascular, neuroendocrine and metabolic responses may all contribute to the sometimes incongruent conclusions from clinical and preclinical studies on the impact of opioids. Nevertheless, the reports in the literature from animal, in vitro and clinical studies urge clinicians to carefully evaluate and further examine the adequacy of selected analgesics used in already haemodynamically and immunocompromised trauma victims. Further understanding of the medical consequences of analgesic and sedative agent selection in the treatment of patients following traumatic injury, sepsis, postsurgically and in the critical care setting is needed to refine the management of the compromised host and in the development of novel approaches for pain alleviation in this population.

Conflict of interest statement

No conflict of interest is declared.


Supported by ONR N00014-97-1-0248.