The severity and frequency of postoperative pain depends on the site, nature and extent of surgery. Despite improvements in the understanding of pain mechanisms and the introduction of acute pain services, the under-treatment of postoperative pain has been recognised as an important issue.[1] The literature indicates that up to 75% of postsurgical patients have reported pain and 80% of these patients experienced severe pain at some time during their hospital stay.[2] A recent survey also demonstrated that approximately 70% of patients still have moderate or severe pain during the perioperative period.[3] Professional care providers have been noted as not believing that patients have pain,[4] and the lack of adequate knowledge and misinformation on pain management among nurses has been identified.[5]

Poorly controlled acute pain can result in increased catabolism, increased cardiorespiratory work, immunosuppression[6] and coagulation disturbances.[7,8] Pain and postoperative nausea and vomiting prolong recovery and discharge times and contribute to unexpected admission after ambulatory surgery.[9] Higher levels of postoperative pain can result in poor patient satisfaction, impair quality of recovery and increase healthcare costs.[10,11] Several clinical studies[12–14] support the notion that an approach based on the use of multimodal techniques to manage postsurgical pain is the most effective strategy for achieving optimal analgesia.

In this article, we provide a review of current strategies for the management of postoperative pain focusing on various analgesics and analgesic techniques. To prepare the article we performed a search of MEDLINE (1980–2006) and the Cochrane library (1980–2006) to identify published reports of randomised controlled trials and systematic reviews. The search terms used were ‘postoperative’, ‘pain’, ‘analgesics’, ‘analgesia’, ‘mechanisms’, ‘opioid’, ‘NSAIDs’, ‘COX-2 inhibitors’, ‘tramadol’, ‘paracetamol’, ‘adjuvants’, ‘regional’, ‘epidural’, ‘intra-articular’, ‘infiltration’, ‘complementary’, ‘acupuncture’, ‘TENS’, ‘music’, ‘hypnosis’ and ‘relaxation’.

1. A Mechanistic Approach to Pain Management

1.1 Mechanisms of Pain

A rational approach to the treatment of pain would be to identify the contributing mechanisms and specifically target treatment (figure 1). Despite recent progress in understanding nociceptive pathophysiology, wide variations in the management of postoperative pain are still observed. This may, in part, be due to the combination of innate individual and surgical factors that contribute to the development of pain. Nociceptive and inflammatory stimuli result in diverse effects, including nociceptive transduction, sensitisation of peripheral nerve endings and central neuronal sensitisation (figure 2).[15] Rarely, a single mechanism may contribute to the pain state; more often, a combination of mechanisms is involved.

Fig. 1
figure 1

Action of analgesics at various sites of the pain pathway. COX = cyclo-oxygenase.

Full size image
Fig. 2
figure 2

Mechanism of peripheral and central sensitisation. BK = bradykinins; LT = leukotrienes; NO = nitric oxide; PG = prostaglandins.

Full size image

This section primarily focuses on the common mediators involved in the transmission and augmentation of noxious stimuli. Most of the nociceptive signals arise from the activation of polymodal nociceptors. These nociceptors have little spontaneous activity under normal conditions but exhibit increased activity, with an increase in response magnitude to noxious stimuli or enhanced receptive field, when tissue injury occurs.[16] A complete review of acute pain mechanisms is beyond the scope of this article.

1.1.1 Peripheral Mechanisms

It is recognised that inflammatory response to tissue injury results in the activation of a cascade of events leading to peripheral and central sensitisation of the nociceptive pathway (figure 2). Noxious stimulation causes selective release of peptides and non-peptides in peripheral tissues, which sensitise the peripheral nerve endings (table I). The source of these substances is varied: injured cells, nociceptors, enhanced capillary permeability and generation by local enzyme activity. These substances are also released in the spinal cord, resulting in sensitisation. For example, substance P is synthesised in the neuronal cell bodies in the dorsal root ganglion and transported to peripheral and central terminals where it is stored in vesicles.[17] It causes vasodilatation, oedema and the release of histamine. Substance P appears to have a role in potentiating both excitatory and inhibitory inputs in the spinal cord, thereby sensitising the neurons to nociceptive signals.[18] It has a slow and prolonged effect during sustained noxious stimulation, binding at neurokinin NK1 receptors and resulting in calcium ion influx into neuronal cells. Substance P also induces the production of nitric oxide, a vasodilator for the endothelium, which, by complex mechanism, exacerbates pain.[19] This finding indicates a potential role for selective substance P antagonists with high affinity for NK1 receptors to prevent the prolongation of nociceptive transmission.[20–22] However, studies to date have not found significant and consistent analgesic properties for NK1 receptor antagonists in humans.[23,24] Serotonin — which is released from non-neuronal cells such as platelets and mast cells and within the digestive tract — is one of the components of the ‘inflammatory soup’ present in the extracellular space of injured tissues. The role of serotonin in nociception is complex. It can cause excitation of nociceptive afferents via the activation of its membrane-bound receptors (5-HT1–3)[25] as well as sensitising nociceptors, especially to bradykinin.[26] Antagonists acting at peripheral serotonin receptors (5-HT2A and 5-HT3) may be of value in reducing pain arising from inflamed tissues.

Table I
figure Tab1

Inflammatory pain mediators

Full size image

1.1.2 Central Mechanisms

In recent years, major developments have been made in our understanding of the role of the spinal cord in nociceptive transmission. Peripheral sensitisation can enhance the pain responses of nociceptive neurons in the CNS. One strategy to reduce pain arising from such a mechanism is to decrease the excitability of these central neurons. There are also a number of peptides involved in augmentation of nociception at the spinal level (figure 2). For example, cholecystokinin (CCK) found in intrinsic spinal neurons is believed to play a role in nociception and selectively reduces the analgesic actions of morphine.[27,28] This indicates that CCK antagonists may enhance opioid analgesia while reducing opioid tolerance.

In vitro studies have shown that both A-beta and C fibre activation increases aspartate and glutamate outflow at the level of the spinal cord,[29,30] resulting in amplification and prolongation of nociceptive signals leading to a chronic pain state (figure 2). The action of these amino acids[31] is mediated via NMDA and non-NMDA receptors. Ketamine (at subanaesthetic doses of 0.15–0.5 mg/kg) and dextromethorphan[32] are clinically effective NMDA receptor antagonists.[33]

There is also a growing body of evidence from animal studies[34–37] which indicates that ascending nociceptive and descending noradrenergic pathways play a critical role in modulation of pain at various levels in the spinal cord. The spinal receptor target for this system is the α2-adrenoceptor. Agonists at this receptor, such as clonidine, produce anti-nociception and may potentiate the actions of morphine.[38]

As pain mechanisms rely on different receptor systems, it is prudent to adopt a multimodal approach to achieve pain relief in the perioperative period.

2. Rationale for Multimodal Analgesia

The ideal postoperative analgesic regimen would provide effective pain relief, reduce opioid-related adverse effects and the surgical stress response, and improve clinical outcome, e.g. morbidity, mortality and duration of hospital stay. The concept of multimodal analgesia was introduced to achieve these goals by combining various analgesic techniques and different classes of drugs to improve postoperative outcome.[39] However, available data are conflicting and do not necessarily show that multimodal analgesia has resulted in improved outcome and concomitant reduction in the adverse effects of opioids.[12,40,41] The failure to improve clinical outcome may be due to inappropriate combination and dosages of analgesics. In addition to adequate analgesia, postoperative morbidity and duration of hospital stay depend on other factors such as initiation of early nutrition, mobilisation and comprehensive rehabilitation programmes.[42] The lack of effect of multimodal analgesia may also be due to inappropriate timing of administration of analgesics. Although there is insufficient evidence to recommend pre-emptive analgesia routinely, it is prudent to attenuate postoperative pain as effectively as possible during the intraoperative period and initiate effective analgesic therapy in the early phase of the perioperative period.

The effectiveness of individual analgesics can be enhanced by combining those acting via different mechanisms, such that additive or synergistic effects are achieved. For example, the synergism between α-adrenergic and opioid systems has been demonstrated, with the finding that clonidine can potentiate the effects of morphine.[43] Similarly, combinations of paracetamol (acetaminophen) and NSAIDs provide additive analgesic effect in mild to moderate acute pain.[44] The addition of cyclo-oxygenase-2 (COX-2) inhibitors or NSAIDs to opioids reduces opioid requirements by 20–30%, with a concomitant reduction in opioid-related adverse effects and better analgesia. For example, in a recent study, patients undergoing total abdominal hysterectomy who were randomised to intravenous administration of parecoxib 40mg at induction of anaesthesia had a significant reduction (by 26%) in 24-hour morphine consumption and lower pain scores on sitting up in comparison with placebo.[45] Furthermore, ketamine, an NMDA receptor antagonist, has been shown to reduce the pain scores and lower analgesic requirement when added to a multimodal epidural analgesia.[46] Adding ketamine to patient-controlled epidural analgesia (PCEA) with morphine, bupivacaine and adrenaline (epinephrine) has been demonstrated to result in an enhanced analgesic effect – the mean visual analogue scale score for pain in the ketamine group during movement and cough were lower than the control group.[47] The cumulative total analgesic consumption in the ketamine group was also lower by 30% than the control group 24 hours following surgery. In another study, it was demonstrated that the combination of intraoperative ketamine and epidural analgesia may confer a long-term benefit in reducing the incidence of chronic pain.[48]

Similarly, transcutaneous electrical nerve stimulation (TENS) administered in an optimal frequency (85Hz) in the wound area can reduce analgesic consumption for postoperative pain. In a meta-analysis, Bjordal et al.[49] reported that TENS resulted in a mean reduction in analgesic consumption by 26% compared with placebo.

In addition to a reduction in total analgesic consumption and better pain scores, reports suggest that multimodal analgesia may confer long-term benefits in patient outcomes.[50–52] Evidence also suggests that inadequate postoperative pain relief may result in significant morbidity that leads to an increase in healthcare costs due to unanticipated hospital admission.[53,54] Epidural analgesia using local anaesthetics and opioids is widely practiced as an important component of a multimodal approach to control postoperative pain and hasten recovery.[55] The synergistic effect of a combination of local anaesthetics and an opioid such as morphine provides superior dynamic analgesia with minimal adverse effects compared with a local anaesthetic or opioid alone.[56] In addition to local anaesthetics and opioids, a number of adjuvants such as ketamine[47] and clonidine[57] have also been used epidurally; however, there are conflicting reports about the value of epidural analgesia in improving outcomes. In a study of patients undergoing abdominal aortic surgery receiving thoracic epidural analgesia, the incidence of thrombotic, infectious and cardiovascular complications as well as the duration of hospital stay was significantly reduced.[58] Furthermore, the combination of epidural analgesia and NSAIDs initiated in the preoperative period has been shown to shorten recovery time.[14] Conversely, in high-risk patients who received perioperative epidural analgesia for major abdominal surgery, Peyton et al.[59] showed no difference in the duration of intensive care or hospital stay, although the findings of this study raise questions about its design and conclusions.[60]

In some patients, the adverse effect profile of various analgesics may affect quality of recovery. Continuous perineural techniques offer the potential benefits of prolonged analgesia, reduced opioid requirements and fewer adverse effects. A prospective evaluation of over 1300 patients undergoing total hip arthroplasty who received continuous peripheral nerve block reported significantly higher satisfaction, fewer adverse effects and less opioid consumption compared with groups that received intravenous patient-controlled analgesia (PCA) with morphine or PCEA.[61]

3. Pharmacological Options and Analgesic Techniques

3.1 Opioids

Opioids are effective analgesics for moderate to severe pain, although their efficacy is limited by adverse effects. They act on opioid receptors in the periphery[62,63] and CNS.

Opioids and/or NSAIDs combined with local anaesthetic infiltration or intra-articular block may be a useful technique for controlling pain in patients after ambulatory surgery. In most studies, local anaesthetic infiltration with systemic opioids or NSAIDs showed improvement in analgesia, better recovery[9] and shortening of discharge time from day surgery unit compared with placebo.[64,65] A prospective 7-year survey of nearly 6000 patients revealed that a high degree of patient satisfaction and low incidence of adverse effects could be achieved with the administration of neuraxial opioid analgesia for major surgery.[66]

Recently, a novel extended-release morphine formulation (morphine in a lyposomal carrier; DepoDur™;Footnote 1 Endo Pharmaceuticals Inc., Chadds Ford, PA, USA) has become available for single-dose epidural use. Several studies in various surgical procedures such as knee replacement, abdominal surgery and caesarean sections have examined the efficacy of this formulation as an alternative to conventional intravenous opioid analgesia.[67–69] Extended-release morphine has been reported to provide analgesia up to 48 hours postoperatively for lower abdominal surgery.[68] Patients treated with 10, 20 and 25mg used significantly less fentanyl via the intravenous PCA system than patients receiving standard epidural morphine. Most frequently reported adverse effects were nausea, pruritus, pyrexia, vomiting and hypotension. In the extended-release morphine group who received 25mg, a decrease from baseline respiratory rate was observed and sedation was more pronounced up to 12 hours postoperatively than in groups that received lower doses. Delayed respiratory depression is the most serious adverse effect of morphine. The authors recommended the optimal dose with fewer adverse effects to be 15mg of extended-release morphine, and that dose reduction is warranted in older patients.

3.2 Paracetamol (Acetaminophen)

Paracetamol is an effective analgesic for mild to moderate pain, with a favourable adverse effect profile.[70] It is an effective adjuvant to opioid analgesia, and a reduction in opioid requirement by 20–30% can be achieved when combined with a regular regimen of oral or rectal paracetamol.[71] It has been shown that 1g of propacetamol results in a significant reduction in postoperative morphine consumption over a 6-hour period.[72] A meta-analysis of analgesic efficacy suggested that paracetamol and tramadol is an effective analgesic combination in dental and postsurgical pain; however, more patients experienced adverse effects such as dizziness, nausea and vomiting with this combination than with either agent alone.[73]

Propacetamol, a prodrug of paracetamol, may be a viable alternative to NSAIDs in the perioperative period in minor surgery.[74]

3.3 NSAIDs

NSAIDs have become the cornerstone in the treatment of acute pain in the early postoperative period because of their opioid-sparing effect.[75] Administration of ibuprofen and oxycodone in combination provides superior and effective analgesia in the postoperative period.[76] The combination of ibuprofen and paracetamol has also been reported to reduce the need for early analgesia by up to 34% in children undergoing tonsillectomy.[77]

3.3.1 Cyclo-Oxygenase-2 Inhibitors

There has been a renewed interest in the use of COX-2 inhibitors for postoperative pain. Because of their favourable adverse effect profile, COX-2 inhibitors provide a safer alternative to conventional NSAIDs. Nonselective NSAIDs are associated with adverse effects related to COX-1 inhibition, which include gastrointestinal ulceration, renal dysfunction and perioperative bleeding.[78] At normal doses, the COX-2 inhibitors selectively inhibit COX-2. In addition, they exhibit an opioid-sparing effect[79] and therefore constitute an important component of multimodal therapy for the treatment of postoperative pain.

Parecoxib (a prodrug of valdecoxib) has analgesic potency similar to that of ketorolac and has been studied extensively in postoperative dental pain and other models.[80–84] Single-dose parecoxib 40mg provided significantly better pain responses compared with placebo or morphine 4mg and was comparable to ketorolac 30mg after gynaecological surgery.[85] The main advantage of parecoxib is that it can be administered intraoperatively and immediately postoperatively before oral medication is tolerated. Gan et al.[79] have shown that preoperative parecoxib followed by postoperative oral valdecoxib reduces opioid requirements after laparoscopic cholecystectomy compared with placebo. The cumulative morphine equivalent dose (MED) of fentanyl in the treated group was lower compared with the placebo group up to 4 days postoperatively. In addition, significantly reduced incidences of opioid-related symptoms were observed in the parecoxib/valdecoxib group.

In another study, patients undergoing knee replacement surgery were randomly assigned to receive another COX-2 inhibitor rofecoxib or placebo perioperatively in conjunction with PCEA. It was shown that the rofecoxib group experienced significantly less breakthrough pain and fewer adverse effects compared with the placebo group. They also experienced quicker recovery based on range of motion and pain scores.[14]

Recently, much attention has focused on the ultimate risk-benefit ratio of COX-2 inhibitors.[86–88] Selective COX-2 inhibitors — by decreasing vasodilatory and anti-aggregatory prostacyclin (epoprostenol) production — may lead to increased prothrombotic activity. By doing this, they may affect the balance between prothrombotic and antithrombotic eicosanoids,[89] resulting in an increased risk of myocardial infarction. However, this theory was considered oversimplified and the underlying mechanism may be more complex.[90–92] As a result of the increased cardiovascular risks, rofecoxib and valdecoxib have been withdrawn from the market. Celecoxib is available for clinical use worldwide, while parecoxib is only available outside the US.

NSAIDs and COX-2 inhibitors are also known to reduce peripheral nociceptor discharge by reducing the local concentration of arachidonic acid metabolites. NSAIDs exhibit central and peripheral analgesic action that varies depending on the type of NSAID used and on the presence or absence of an inflammatory process.[93] Romsing et al.[94] investigated the evidence for a peripheral analgesic effect of local infiltration with NSAIDs. Five studies compared intra-wound infiltration of tenoxicam (7.5 or 10mg) or ketorolac (30 or 60mg) with similar systemic (intravenous or intramuscular) administration in patients undergoing inguinal hernia repair or mastectomy. Two of the five trials showed significantly lower pain scores at 2 hours and 24 hours after wound infiltration with NSAIDs. The 24-hour consumption of supplementary analgesics was significantly reduced by approximately 60% in the wound infiltration group. It should be emphasised that the observed benefit may be due to a systemic effect of the locally administered NSAID.

3.4 Tramadol

Tramadol is a synthetic, centrally acting analgesic with weak agonist activity at opioid receptors. It also inhibits serotonin and noradrenaline (norepinephrine) reuptake.[95] The tramadol metabolite, O-desmethyl tramadol, is a more potent analgesic than tramadol.[96] Unlike other opioids, tramadol lacks respiratory depressant effects and exhibits a lower risk of bowel dysfunction[97] at conventional doses.

Moore and McQuay,[98] in a meta-analysis, compared single oral doses of tramadol alone with a combination of standard analgesics (aspirin/codeine and paracetamol/propoxyphene) in post-surgical and dental pain. Tramadol 100mg and 150mg showed a number-needed-to-treat value of 4.8 and 2.4, respectively, compared with 3.6–4.0 with combination analgesics. Frequencies of adverse effects were higher with increasing doses of tramadol.

A meta-analysis[99] of a paracetamol and tramadol combination confirmed superior analgesia without additional toxicity. The most common adverse effects noted were diziness, headache, nausea and vomiting. Another study noted that the combination of tramadol with paracetamol increased the tolerability of tramadol.[100]

3.5 Ketamine

There has been renewed interest in the role of ketamine in enhancing postoperative analgesia. Several studies have focused on demonstrating the use of subanaesthetic doses of ketamine for various surgical procedures to enhance pain relief and reduce total analgesic consumption.[101–106] Central excitatory neurotransmitters acting on NMDA receptors have been identified to be involved in the development and perpetuation of pathological pain states causing hyperalgesia and allodynia.[107] Ketamine acts as an antagonist at the NMDA receptor.

There is much evidence (table II) to suggest that ketamine has an opioid-sparing effect and may confer advantages in patients in whom high postoperative opioid consumption is anticipated. At low doses, ketamine can provide improved analgesia in opioid-resistant pain.[108] Continuous infusion of ketamine has been used perioperatively. Adam et al.[109] evaluated intravenous ketamine with an initial bolus (0.5 mg/kg) followed by continuous infusion of 3 μg/kg/min intraoperatively in combination with continuous femoral nerve block in patients undergoing total knee arthroplasty. In this multimodal approach, the ketamine group required significantly less morphine than patients receiving only femoral nerve block and tolerated early mobilisation of the knee.[109]

Table II
figure Tab2

Studies on perioperative use of ketamine (KET)

Full size image

A systematic review[113] of randomised double-blinded trials involving epidural ketamine added to various opioid-based regimens suggested improved analgesia with no increase in ketamine-associated adverse effects. Ketamine in a multimodal regimen of PCEA has been shown to reduce analgesic requirement after major surgery.[47,112] Wong et al.[114] demonstrated that the addition of ketamine to epidural morphine potentiates the analgesic effect of morphine in patients undergoing total knee replacement. Similarly, Chia et al.[47] showed that coadministration of a small dose of ketamine 0.4 mg/mL with a multimodal regimen (morphine, bupivacaine and adrenaline) via PCEA provides better pain relief at rest, during coughing or movement, and a reduction in analgesic consumption compared with PCEA alone. Recently, transdermal ketamine has been used as an adjuvant to epidural analgesia.[115] Patients who were treated with a ketamine patch after minor gynaecological surgery showed prolonged time to first rescue analgesia.

In essence, ketamine has found a role in the management of acute pain but its use as an analgesic has been limited by its unpleasant and unpredictable adverse effects; however, lower doses of ketamine have not been found to be associated with these effects.[116]

3.6 Clonidine

The α2-adrenoceptor agonists, clonidine and dexmedetomidine,[117] have anti-nociceptive activity via peripheral, supraspinal and primarily spinal cord mechanisms including activation of postsynaptic α2-adreoceptors of descending noradrenergic pathways.

Epidural clonidine has several potential advantages over epidurally administered local anaesthetics and opioids because it is devoid of some of the adverse effects associated with these drugs such as motor block, urinary retention, respiratory depression and pruritus. In addition, clonidine exhibits a synergistic action when used as an adjuvant to local anaesthetics or opioids, resulting in a reduction in postoperative analgesic requirement.[57,118] Paech et al.[119] have demonstrated that postoperative epidural analgesia (with bupivacaine and fentanyl) is enhanced significantly by the addition of clonidine. They reported a reduction in opioid requirement and lower pain scores on coughing.

The adverse effects of clonidine are dose related. When tested in healthy volunteers in a dose of 700μg epidurally, Eisenach et al.[120] found that clonidine significantly reduced pain induced by ice water immersion, decreased plasma noradrenaline levels and caused haemodynamic changes, but resulted in intense sedation that lasted up to 6 hours.

Dexmedetomidine has also been used for sedation in intensive care[121] and as an analgesic adjunct during the intraoperative period. It has been demonstrated to provide an opioid-sparing effect with minimal adverse effects.[122,123]

3.7 Epidural Analgesia

Epidural analgesia provides superior pain relief and attenuates the stress response to surgery and pain, particularly when used in the form of continuous infusion both during and after surgery. However, clinical opinion remains divided regarding the benefits of epidural analgesia despite several studies demonstrating benefits such as greater patient satisfaction, less postoperative morbidity and improved clinical outcome.[55,59,124–133] The results of some of the more recent studies, meta-analyses and reviews are summarised in table III. A recent meta-analysis confirmed the value of epidural analgesia in reducing cardiac, pulmonary, thromboembolic and renal complications, in addition to providing superior postoperative analgesia.[126]

Table III
figure Tab3

Studies on epidural analgesia in the perioperative period

Full size image

Combined use of epidural local anaesthetics and adjuvants not only provides intraoperative analgesia but can also control postoperative pain effectively after major thoracic, abdominal and lower limb surgeries.[135] Using adjuvants in the epidural space reduces the total dose of local anaesthetic. In children, several adjuvants have been used in caudal blocks, including ketamine and clonidine, for prolongation of analgesia after common paediatric operations.[136]

A study conducted in patients undergoing major intra-abdominal surgery demonstrated better pain relief and reduced postoperative analgesic consumption with epidural analgesia.[133] For hip and knee replacements, epidural anaesthesia with sedation during surgery followed by postoperative epidural infusion of a combination of local anaesthetics and opioids has been used.[137] Lee et al.[138] demonstrated a clear advantage of the combination of epidural bupivacaine and diamorphine for analgesia after major gynaecological surgery. The combination provided superior pain relief with fewer adverse effects.

Nonetheless, there are problems associated with the continuous epidural technique such as hypotension, motor blockade and incompatibility with anticoagulation. In addition, the failure rate with epidural analgesia is approximately 30%.[139]

3.8 Wound Infiltration with Local Anaesthetics

Infiltration of the surgical wound with local anaesthetics is commonly performed to achieve wound analgesia. This method has been shown to be effective in providing analgesia in trials where patients have undergone inguinal hernia repair, although there is lack of evidence for any clinically useful effect for most other abdominal procedures (see review by Moiniche et al.[140]). Inadequate doses of local anaesthetics may explain the poor results in some trials. Wound infiltration does not provide a beneficial effect on pulmonary function after surgery.[129]

The role of adjuvants to local anaesthetics is unclear. For example, wound infiltration with bupivacaine and ketamine has not been shown to decrease pain score or the need for rescue analgesia, but the duration of analgesia has been reported to be prolonged by the addition of ketamine.[141] The routine use of adjuvants in wound infiltration is currently not recommended.

3.9 Intra-Articular Analgesics

There are conflicting reports about the efficacy of intra-articular analgesics.[142–144] Intra-articular NSAIDs when used alone are unlikely to exert any clinically useful analgesic effect, but NSAIDs in combination with local anaesthetics may provide prolonged analgesia.[145] Many orthopaedic surgeons use intra-articular local anaesthetics following arthroscopic surgery. In a systematic review, Moiniche et al.[146] observed that the use of intra-articular local anaesthetics seemed to provide moderate pain relief of short duration. Although the pain relief in the early postoperative period was statistically significant, evidence was not overwhelming in favour of intra-articular local anaesthetics because the majority of the studies did not demonstrate improved pain relief beyond the immediate postoperative period.

Peripheral anti-nociceptive actions of opioids have been reported,[147] and thus opioids appear to be effective in pain control when administered intra-articularly. Stein[148] showed that morphine exerts its effect on peripheral opioid receptors when injected intra-articularly; however, a systemic effect cannot be excluded. Gupta et al.,[149] in a meta-analysis, evaluated 27 studies where intra-articular morphine was directly compared with placebo. In 13 of the studies, intra-articular morphine had a mild beneficial effect (mean reduction in pain intensity, 12–17mm on a visual analogue score). Several studies reported a reduction in analgesic requirement in the morphine group. The efficacy of morphine depends on the dose used. A higher dose (5mg) injected intra-articularly is likely to provide analgesia during the first 24 hours after surgery, whereas lower doses are ineffective.

Romsing et al.[94] in a systemic review highlighted four studies that compared intra-articular ketorolac 60mg or tenoxicam 20mg with systemic (intravenous) administration. In two of these ketorolac reports intra-articular bupivacaine was added in both groups.[145,150] All four studies showed significantly lower pain scores after intra-articular administration compared with systemic administration of NSAIDs. In the intra-articular groups, time to first analgesic request was significantly increased postoperatively.

Data also suggest that using a combination of analgesics intra-articularly has beneficial effects. Patients who received intra-articular ketorolac in addition to morphine demonstrated significantly lower pain scores compared with patients receiving either drug alone postoperatively.[151] Similarly, a combination of clonidine and neostigmine[152,153] have also been shown to exhibit peripheral analgesic effects when used intra-articularly after knee arthroscopy. At a dose of 150μg intra-articularly, clonidine exerted analgesia comparable to morphine but the combination of both did not provide prolonged analgesia.[154]

3.10 Peripheral Nerve Blocks

Appropriate nerve blocks, depending on the site of surgery, are useful in providing short-term pain relief post-operatively. Direct visualisation of neural tissue with ultrasound technology and the utility of stimulating catheters has made placement of indwelling catheters safer and accurate.

Continuous infusion of local anaesthetics through a peripheral nerve catheter is becoming increasingly popular in both hospital and ambulatory settings to achieve prolonged analgesia.[155,156] For example, continuous femoral nerve block has been shown to reduce the duration of hospital stay and the frequency of serious complications after total knee arthroscopy.[157] Similarly, several other studies have demonstrated the benefits of peripheral nerve blocks (PNBs), including reduced duration of stay and costs,[155] decreased incidence of postoperative nausea and vomiting,[158] and lower rates of unexpected hospital admissions after ambulatory surgery.[155,157,159]

In a recent meta-analysis involving over 600 patients, PNBs provided superior postoperative analgesia when compared with administration of oral and systemic opioids alone.[160] Patients who received perineural blocks with local anaesthetic also demonstrated a significant reduction in opioid consumption, fewer opioid-related adverse effects and better patient satisfaction. Of the six studies that evaluated clonidine,[161] five found improvement in analgesia.

Evidence that the addition of peripherally administered opioids during PNB[162,163] improves the quality of either intraoperative regional anaesthesia or postoperative analgesia has not been established, with mixed results from different trials. Five trials measuring postoperative efficacy reported a significant difference in favour of the opioid;[162] however, the authors did not consider the findings important enough to advocate routine use of opioids in the acute setting.

3.11 Pre-Emptive Analgesia

With a greater understanding of acute pain mechanisms, it is known that tissue injury initiates peripheral and central neuronal sensitisation, resulting in perpetuation of the ‘pain state’.[164–166] Pre-emptive analgesia is analgesia given before the initiation of nociceptive stimulus. Several clinical studies hypothesised that preoperative administration of analgesics or regional blockade would prevent or result in less postoperative pain[106,167–170] by protecting the nervous system from sensitisation. However, the effectiveness of pre-emptive analgesia in routine clinical practice is debatable.[171,172] For pre-emptive analgesia to be clinically important the duration of analgesia obtained should be longer and with a minimal increase in adverse effects, and the pre-emptive intervention should have an effect on acute postoperative pain or in preventing the development of chronic pain.

A recent review of the literature found that around 40% of studies demonstrated a beneficial effect (reduction in pain and analgesic consumption) of pre-incision (pre-emptive) analgesia as opposed to analgesic administration after surgical incision (preventive).[173] Therefore, along with other factors (drug, dose and duration) it appears that the timing of analgesic initiation in relation to surgical incision affects postoperative analgesic requirement when analgesic activity of the agent has worn off.[174,175] The timing of administration of an agent may be pre-incision,[176] during surgery[177] or in the postoperative period.[178]

NMDA antagonists,[172] NSAIDs,[179] epidural analgesia[180] and local anaesthetic infiltration[170,181,182] have all been used in pre-emptive analgesia, with variable results.

4. Novel Analgesic Therapy

4.1 Gabapentin

Gabapentin, which primarily has anticonvulsant properties, is used extensively in the treatment of neuropathic pain. Despite its structural similarity to GABA, gabapentin does not bind to GABA receptors.[183] It has high affinity for α2-δ subunits of voltage-dependent calcium channels, resulting in postsynaptic inhibition of calcium influx and thereby reducing presynaptic excitatory neurotransmitter release.[184] It has been suggested that gabapentin is useful in reducing the central neuronal sensitisation that occurs in postoperative pain, and postsurgical pain has been regarded by some as transient neuropathic pain.[185]

Numerous studies have demonstrated the efficacy of gabapentin, and another GABA analogue pregabalin, as non-opioid analgesic adjuvants in postoperative pain management.[50,186–196] Table IV summarises some of these studies.

Table IV
figure Tab4

Studies on perioperative use of gabapentin (GAB) and pregabalin (PGB)

Full size image

In a randomised study using single-dose gabapentin (1200mg) versus placebo, Dirks et al.[192] found that there was a substantial reduction in morphine consumption after mastectomy in the gabapentin group (total morphine consumption, 29mg in the placebo group versus 15mg in the gabapentin group). In addition, pain scores were lower in the treatment group in the early postoperative period. Dierking et al.[193] also demonstrated a reduction in morphine consumption (by 32%) when gabapentin (3000mg) was administered before and during the first 24 hours after abdominal hysterectomy. A recent meta-analysis confirmed that gabapentin in doses <1200mg has an analgesic and opioid-sparing effect in acute postoperative pain management when used in conjunction with opioids; however, it was associated with an increased risk of sedation but reduced opioid-related adverse effects such as vomiting.[194] In a multi-modal approach, Gilron et al.[195] reported that the combination of gabapentin and rofecoxib is superior to either agent alone after abdominal hysterectomy. With all patients receiving intravenous PCA, the gabapentin-rofecoxib combination (1800/50mg) demonstrated a significant reduction in morphine consumption, from 130mg in the placebo group to 57mg in the combination group.

Based on the current evidence it appears that gabapentin reduces supplementary postoperative analgesic requirement. The optimal dose ratio devoid of adverse effects needs to be identified when gabapentin is used alone or in combination.

4.2 Corticosteroids

Corticosteroids have been used as anti-inflammatory and anti-immunological agents for manifestations of many autoimmune disorders. Corticosteroids act by suppressing arachidonic acid production through lipocortin-induced phospholipase inhibition, which ultimately inhibits production of both prostaglandins and leukotrienes.[198] Corticosteroids also prevent the production of cytokines, which play a role in the mechanism of inflammatory pain. At the spinal level, corticosteroids exert anti-nociceptive effects.[199]

Despite evidence for analgesic actions of corticosteroids, there has not been widespread clinical use of these agents for the management of postoperative pain. This may be due to the adverse effects of corticosteroids when given in repeated doses for longer periods postoperatively; however, most studies have reported their analgesic effect after single doses. Several clinical investigations have evaluated the effect of systemic corticosteroids on pain after surgical procedures, demonstrating effective pain relief and early recovery.[200–202] In a double-blind, placebo-controlled, single-dose, randomised study, Romundstad et al.[203] compared intravenous ketorolac with intravenous methylprednisolone 125mg in patients with moderate to severe pain one day after orthopaedic surgery. They demonstrated that the pain intensity in the corticosteroid group was significantly lower up to 6 hours postoperatively compared with placebo, and similar to ketorolac. Opioid consumption in the first 72 hours was significantly lower in the methylprednisolone group than in the ketorolac and placebo groups. No serious adverse effects were reported.

Glucocorticoids may have sustained analgesic effects after surgery; however, the disadvantage of repeated corticosteroid administration and adverse effects in postoperative patients needs further investigation in adequately powered trials.

4.3 Opioid Receptor Antagonists

Naloxone and nalbuphine (a partial opioid receptor agonist) are used to treat neuropathic pain.[204] In vitro studies have demonstrated that a long-acting nalbuphine preparation had antinociceptive action for up to 55 hours.[205] Evidence from animal studies also illustrates that opioid antagonists enhance the potency of morphine and attenuate opioid tolerance.[206,207]

In one study, low-dose naloxone infusion in the postoperative period after gynaecological surgery was shown to reduce opioid-related adverse effects and opioid requirements.[208] However, there is inconclusive evidence from several other clinical trials on the effectiveness of opioid antagonists to attenuate postoperative pain and reduce opioid requirements.[209,210] In a recent randomised, double-blinded clinical study, Cepeda et al.[211] compared a naloxone-morphine combination with morphine alone in PCA. There was no difference in opioid requirements and pain between the groups. The combination group experienced less nausea and pruritus than the morphine only group; however, naloxone was unable to reduce the vomiting, sedation and urinary retention seen in the morphine only group.

4.4 Magnesium Sulfate

Recently, the role of magnesium sulfate as an NMDA receptor antagonist and as an adjuvant to analgesic therapy has been investigated. Intravenous magnesium sulfate 50 mg/kg preoperatively followed by an infusion postoperatively in patients undergoing open cholecystectomy has been reported to result in less discomfort than saline treatment,[212] although there was no effect on opioid requirement. In another study,[213] patients randomised to 20% intravenous magnesium sulfate before surgery showed reductions in postoperative PCA morphine requirements in the first 48 hours compared with patients receiving saline.

Contrary to the above findings, in a systematic review examining the efficacy of NMDA antagonists, McCartney et al.[214] reported that none of the four studies examining magnesium to reduce postoperative pain and analgesic requirements demonstrated any evidence in favour of preventive analgesia; in contrast, ketamine and dextromethorphan were effective in this regard.

4.5 Lidocaine (Lignocaine) Patches

Lidocaine (lignocaine) provides analgesia by blocking the sodium neuronal channels locally and thereby dampens peripheral nociceptor sensitisation. Transdermal lidocaine (5%) has been used to treat pain associated with post-herpetic neuralgia.[215] When lidocaine is applied as a topical agent, its serum concentrations are insignificant.[216] There may be a potential role of topical lidocaine in acute postoperative pain but this needs further investigation.

4.6 Patient-Controlled Transdermal Fentanyl

Patient-controlled transdermal fentanyl offers a non-invasive opioid delivery system for acute pain management. Using iontophoresis technology, fentanyl 40μg is delivered on demand.[217] A multicentre trial reported that patient-controlled transdermal fentanyl was superior to placebo in controlling moderate to severe pain up to 24 hours after major surgery.[218]

5. Nonpharmacological Options

Nonpharmacological therapy should be considered as complimentary to pharmacological options for postoperative pain management. It provides additional benefit in reducing the total dose of analgesics required and therefore minimising the adverse effects of the analgesics.

5.1 Transcutaneous Electrical Nerve Stimulation

TENS has been used widely in chronic pain conditions. The evidence for its efficacy in acute postoperative pain is inconclusive, mainly due to a lack of well-conducted, randomised controlled trials. Many published studies have methodological flaws. Blinding is difficult in trials associated with TENS because patients easily notice the presence or absence of paraesthesia; therefore, inadequate randomisation may exaggerate the efficacy by up to 40%.[219]

A systematic review by Carroll et al.[220] demonstrated TENS to be no better than placebo in the treatment of acute postoperative pain. A wide variety of procedures from hernia to thoracotomy were studied and ten different TENS units were used with different control settings and duration of treatment. Fourteen trials compared TENS with sham TENS. None found any difference between the two treatments. However, when TENS was compared with opioids, one trial[221] reported significantly fewer pethidine injections and better pain scores on the first postoperative day in patients receiving TENS.

Previous systematic reviews[49,222] on TENS and postoperative pain showed several inconsistencies in the effectiveness of TENS. All available trials of postoperative pain used TENS as an adjuvant to medication, and it is possible that the effect of TENS was masked by the analgesic effect of the medication.

5.2 Relaxation Techniques

Relaxation has become increasingly popular as an adjuvant to conventional analgesic therapy. It has been suggested that it works by breaking the pain-tension-pain cycle.[223]

Seers and Carroll[224] conducted a systematic review to investigate the effectiveness of relaxation techniques when used alone in the management of acute pain. Three of the seven studies they assessed demonstrated significantly less pain or pain distress with relaxation. Most studies were poorly designed with a lack of adequate controls, and a variety of relaxation techniques was used. This meta-analysis proposed the need for better quality trials to establish the efficacy of relaxation techniques for acute postoperative pain before it is widely accepted as a routine analgesic intervention.

5.3 Music Therapy

Music therapy during and after surgery has been used as complimentary to other methods of pain management. Several studies claim benefit of music therapy in reducing pain scores and anxiety.[225–227] Nilsson et al.[226] in a controlled trial examined 151 patients undergoing day case surgery for inguinal hernia repair or varicose veins surgery under general anaesthesia. Patients who were exposed to music intraoperatively or postoperatively reported lower pain intensity in the early postoperative period than a control group who were exposed to ‘white noise’. It is interesting to note that the postoperative music group required less morphine at 1 hour compared with the control group in the recovery room. The pain scores at home on the day of surgery and up to 48 hours afterwards were low and there were no significant differences between the three groups. This study suggests that music therapy has a short-term benefit in reducing pain and anxiety, decreasing the pain perception through cognitive coping.[227] The type and duration of music therapy needed are still unclear. However, a recent systematic review demonstrated a small reduction in pain scores and opioid consumption after music therapy.[228]

5.4 Acupuncture

Acupuncture, a form of traditional Chinese medicine, has been a topic of interest in managing postoperative pain. Several studies of acupuncture have demonstrated an analgesic effect and a reduction in the incidence of nausea and vomiting.[229,230] The success of acupuncture as an adjuvant in balanced analgesia depends on several factors, including the skills of the acupuncturist,[231] method of stimulation and duration of acupuncture therapy. The mechanism by which acupuncture produces analgesia is still unclear. The ‘gate control theory’ and secretion of endogenous opioids such as endorphins, encephalins and dynorphins may contribute to acupuncture-induced analgesia.[232,233]

Kotani et al.[234] tested the effect of acupuncture on postoperative pain in a controlled and double-blind study involving patients undergoing abdominal surgery. The investigators preoperatively inserted intradermal needles at acupoints 2.5cm bilaterally from the spinal vertebrae (bladder meridian). All these patients had an epidural catheter inserted for abdominal surgery. The pain relief was significantly better in the treatment group than in the control group until the second postoperative day. The treatment group required less morphine, by up to 50%. In contrast, Sim et al.[235] have shown that preoperative electro-acupuncture-induced analgesia did not reduce 24-hour morphine requirements after gynaecological lower abdominal surgery. This may be due to the short duration of action of electro-acupuncture when used during the preoperative period.

5.5 Hypnosis

Numerous studies have demonstrated the efficacy of hypnosis for reducing pain in the laboratory setting and case reports have indicated reductions in clinical pain.[236–239] There is evidence to suggest that hypnosis may modify pain perception, at least to some degree, through inhibition at the spinal level.

In acute pain, there is a substantial amount of anecdotal evidence and some controlled studies to support the efficacy and use of hypnosis. Montgomery et al.[240] conducted a meta-analysis to estimate the effectiveness of adjunctive hypnosis in controlling signs and symptoms after surgery. This meta-analysis suggested that an average 89% of surgical patients in hypnosis groups benefited relative to control patients. They also observed that the patients in the treatment group reflected greater satisfaction than the controls.

Evidence suggests that hypnosis is a useful tool to modulate pain and to alter a patient’s perception to change their expectations about pain;[241] however, not all patients necessarily exhibit similar responses to hypnotic suggestions.[242]

6. Conclusion

The management of acute postoperative pain continues to pose a major challenge for healthcare providers. Advances in understanding of the mechanisms of pain, including the receptors involved in the transmission of pain, have led to improvements in postoperative pain management. Nevertheless, available data indicate that acute pain is under-treated.

It has been demonstrated that patients who are exposed to multimodal pain therapy experience less postoperative complications and a reduced duration of hospital stay, indicating that a combination of modalities will result in an improvement in postoperative pain and better clinical outcomes. The analgesic techniques used should be individualised to the patient and the type of surgical procedure. For minor surgeries, a combination of NSAIDs, paracetamol and local anaesthetic infiltration may be sufficient to provide analgesia. For patients in whom severe pain is anticipated, a combination of paracetamol, NSAIDs, systemic opioids or continuous epidural analgesia with local anaesthetic and opioid combination should be considered. PNBs are likely to be more effective for limb surgeries. Evidence suggests that acute pain can trigger long-term plastic neuronal changes leading to a chronic pain state. Drugs such as ketamine and gabapentin given in the perioperative period may reduce the incidence of chronic pain. Combination analgesics supplemented by nonpharmacological therapy will continue to play a vital role in the everyday comprehensive management of acute postoperative pain.