REVIEW URRENT C OPINION
Nonopioid analgesics for postoperative pain management Esther Pogatzki-Zahn a, Chandani Chandrasena b, and Stephan A. Schug b,c
Purpose of review Postoperative pain remains poorly treated in many patients. One reason is the inadequate use of nonopioid analgesics. This review examines the most recent findings on nonopioid analgesics and how these translate into clinical practice. Recent findings Commonly used nonopioid analgesics are paracetamol, NSAIDs and metamizol (dipyrone). For paracetamol, the efficacy is obviously inferior to NSAIDs and the risk of adverse events might have been underestimated in the past. For NSAIDs, there are increasing data supporting similar efficacy of nonselective and cyclooxygenase-2 selective NSAIDs, although the adverse effects of the latter might make them the preferred option for short-term use in the perioperative setting. Metamizol (dipyrone) is a very effective nonopioid analgesic not worldwide available; the discussion on the relevance of the very rare adverse event agranulocytosis continues. Summary Nonopioid analgesics are important components of multimodal postoperative analgesia. The selection of the most appropriate compound for an individual patient can be based more and more on ever increasing data on these important analgesics. Keywords cyclooxygenase inhibitor, dipyrone, multimodal analgesia, NSAIDs, paracetamol, postoperative pain
INTRODUCTION Postoperative pain is still poorly managed mainly because therapeutic options are not used appropriately. In fact, more minor surgical procedures seem to be associated with considerable pain indicating that evidence-based treatment recommendations might not be used in routine clinical practice [1 ]. In parallel, there has been an increasing emphasis on providing good postoperative analgesia not only for humane reasons but also to improve postoperative morbidity, possibly mortality, return to function, improved rehabilitation and earlier discharge [2]. In this context, recent studies show an impact of postoperative pain on cognitive function of aging animals, including the development of memory deficits [3]. Thus, prevention of postoperative cognitive deficits in elderly patients might be another result of improved postoperative pain management. Multimodal analgesia, that is combining drugs with different mechanism of action [such as opioids together with nonopioid analgesics (NOPAs)] [4 ], is recommended as a best practice in almost every guideline on postoperative pain [5]. However, it is &
&
still a matter of debate which nonopioid analgesic is best suited for the perioperative period. On the one hand, efficacy is very different between some NOPAs. On the other hand, like other medications, NOPAs are not without side-effects and there is increasing awareness of their potential harm. Balancing efficacy against side-effects should determine the choice of the ‘right’ analgesic in each individual patient. The following review illustrates and discusses the increasing evidence of efficacy, side-effects,
a Clinic of Anaesthesiology, Intensive Care Medicine and Pain Therapy, Muenster, Germany, bDepartment of Anaesthesia and Pain Medicine, Royal Perth Hospital and cSchool of Medicine and Pharmacology, University of Western Australia, Perth, Australia
Correspondence to Esther Pogatzki-Zahn, MD, Department of Anesthesiology, Intensive Care Medicine and Pain Therapy, University Hospital Muenster, Albert-Schweitzer-Campus 1, A1, 48149 Muenster, Germany. Tel: +49 251 8347261; fax: +49 251 88704; e-mail: [email protected] Curr Opin Anesthesiol 2014, 27:513–519 DOI:10.1097/ACO.0000000000000113
0952-7907 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-anesthesiology.com
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Pain medicine
KEY POINTS Nonopioid analgesics are an important component of multimodal perioperative analgesia. Paracetamol is a weak nonopioid analgesic with a risk of potential adverse events that might have been underestimated in the past. Nonselective NSAIDs and selective COX-2 inhibitors are similarly effective in reducing postoperative pain, have an opioid-sparing effect and reduce opioid-related side-effects; the most relevant side-effects of nonselective NSAIDs are related to the gastrointestinal tract and are reduced if selective COX-2 inhibitors are used. Cardiovascular risk factors are associated with both nonselective NSAIDs and selective COX-2 inhibitors in patients at risk. Patients with low cardiovascular risk can be safely treated perioperatively with nonselective NSAIDs as well as COX-2 inhibitors (except for cardiac surgery, in which all NSAIDs are contraindicated). Dipyrone (metamizol) has similar efficacy as NSAIDs, but a better side-effect and risk profile. Most common side-effects are reduction of blood pressure and allergic reaction, but the incidence of the feared drug-induced agranulocytosis is very rare. Therefore, dipyrone remains a commonly used nonopioid analgesic for postoperative pain management in many countries including Germany, where it is most commonly used.
indications and contraindications of nonopioids in the postoperative setting by analysis of the most recent literature.
PARACETAMOL (ACETAMINOPHEN) Paracetamol is frequently used in the perioperative period. Although the mechanism of action remains unclear, it is becoming increasingly obvious that this is a central one [6 ]. &
Efficacy of paracetamol The debate on the value and use of paracetamol as a component of multimodal analgesia continues; reports on the efficacy of paracetamol on pain after surgery are variable and sometimes rather disappointing. A recent large and very comprehensive meta-analysis [7] clearly points toward a weak effect of paracetamol compared to NSAIDs with a numberneeded-to-treat (NNT) of 3.5 when 1000 mg are used and lower paracetamol doses being less effective (NNT 4.3 with 600/650 mg). In accordance, the opioid-sparing effect of paracetamol is variable and with only very little or no effect on opioid-related side-effects [8]. In a recent systematic review in 514
www.co-anesthesiology.com
orthopedic surgery [9], six out of nine trials found better analgesia with paracetamol compared with placebo whereas three found no benefit here. A recent systematic review in children found less positive studies with paracetamol than with NSAIDs [10 ]; this was influenced by dose and route of administration. Comparative reviews against ketorolac or ibuprofen in children (and adults) indicate comparable [11] or inferior [12] analgesic potency. &&
Routes of administration for paracetamol As rectal application has an unpredictable bioavailability and should be avoided, intravenous (IV) paracetamol might be an alternative option if the oral route is not available [13]. Overall efficacy of IV paracetamol seems to be similar or – if the duration of the effect is taken into account – even less compared with that of oral paracetamol [14]. As shown by this meta-analysis, the NNT for IV paracetamol and IV propacetamol (a prodrug of paracetamol) is 4.0 at 4 h and decreases to 5.3 at 6.0 h; similarly, the authors found a 30% reduction in opioid requirements compared with placebo over the first 4 h after IV paracetamol and only a 16% reduction over the 6 h after IV paracetamol [14]. In view of the significant costs compared with oral paracetamol and potential side-effects of IV infusion including a need for venous access, IV paracetamol should only be used in patients who are not able to take oral paracetamol and only if more effective NOPAs are not indicated because of patient-related risk factors; further studies to evaluate the utility [15] and the cost-effectiveness [16] of this preparation are needed. This discussion might be influenced in favour of IV paracetamol by a meta-analysis, which showed significant reduction of postoperative nausea and vomiting by prophylactically administered IV paracetamol [17 ]. The effect was best with preoperative prophylactic administration [relative risk 0.54 (0.40–0.74) with less benefit when used before arrival in the recovery room 0.67 (0.55–0.83)] and none when used after onset of pain [1.12 (0.85– 1.48)]. It is of note that a reduction of postoperative nausea and vomiting (PONV) correlated significantly with the reduction of pain and not with reduced opioid use related to the paracetamol application [17 ]. The authors suggest either a direct link between analgesic effect and PONV reduction or a direct effect of paracetamol on PONV. The combination of paracetamol and NSAIDs has previously been shown to be superior in efficacy to paracetamol alone, but not NSAID alone [18]. A recent Cochrane Review [19] now indicates specifically for paracetamol and ibuprofen better analgesia and reduced adverse effects with the combined use &&
&&
Volume 27 Number 5 October 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Nonopioid analgesics for pain management Pogatzki-Zahn et al.
compared to ibuprofen alone. However, results were based on two studies from the same study group and need further confirmation.
Side-effects and risk of paracetamol One reason for the widespread use of paracetamol worldwide is the belief that paracetamol is a harmless substance with rare systemic side-effects. This belief is increasingly challenged by more recent findings [20,21]. Ongoing discussions focus on hepatotoxicity not only because of overdose, but even with therapeutic doses [22]. Liver failure due to paracetamol intake is the most common cause of acute liver failure; the increasing number of unintentional liver failure is worrying, including nonoverdose paracetamol toxicity with a three times higher rate than all NSAIDs pooled [23]. Furthermore, the therapeutic window is narrow in adults and even less in children, and as conditions of metabolism can be altered by various factors, it seems almost impossible to predict an enhanced risk in individual patients [24 ]. The association between paracetamol intake and an increased risk of allergic asthma, rhinoconjunctivitis and eczema during the intrauterine period, infancy and adult life is becoming more obvious. Although a causal relationship between paracetamol and asthma in children after exposure in pregnancy or use in early infancy has not been established yet, there is now overwhelming epidemiological data supporting this association [25]. Although there remains a potential influence of unknown confounding factors, the findings are not affected by identified potential confounding factors. The authors of the above review call the evidence ‘compelling’ and demand randomized controlled trials and establishment of causation. This is supported by other recent literature including a review, which even asks if pediatric use should be banned [26]. This study offers also a review of the biological plausibility of an effect of paracetamol on airway reactivity via increased oxidant-induced airway inflammation due to glutathione effects. This is confirmed by the finding that the association of the risk is increased in children genetically susceptible due to antioxidant genes [27 ]. Antibiotics might add to this effect as suggested by other authors [28]. The increased risk of asthma is even linked to intrauterine exposure to paracetamol by intake of the mother [29 ]. More recently, it has been suggested that paracetamol is associated with an increased risk of hyperkinetic disorders, autisms and kryptorchismus in children whose mothers have taken paracetamol during pregnancy [30,31]. &
&&
&
Cardiovascular effects of paracetamol A potential cyclooxygenase-2 (COX-2) inhibitory effect of paracetamol suggests possible cardiovascular side-effects of paracetamol similar to those associated with NSAIDs (see below), at least in patients with risk factors [21]. However, the data on blood pressure increase in patients after exposure remain unclear; a recent systematic review confirmed this uncertainty [32]. The clinical relevance of this needs to be determined, but it is possibly not relevant for short-term use in the postoperative setting.
NSAIDS The effect of NSAIDs is related to the inhibition of cyclooxygenases.
Efficacy of NSAIDs (nonselective and cyclooxygenase-2-specific NSAIDs) There is no question about the efficacy of nonselective NSAIDs and selective COX-inhibitors in reducing acute postoperative pain, opioid requirements and opioid-related adverse events in the perioperative period. Compared with paracetamol, NSAIDs are more effective in reducing postoperative pain [7]. For instance, the NNTs are for ibuprofen 400 mg 2.5, naproxen 2.7 [7], celecoxib 400 mg 2.6 [33] and etoricoxib 90 mg 1.7 [34]. In addition, recent analyses confirm earlier results showing that NSAIDs are able to reduce opioid requirements in postoperative patients with the significant benefit of reduced side-effects of opioids, such as nausea, vomiting and sedation [35,36 ]. All available data suggest similar efficacy of nonselective NSAIDs and selective COX-2 inhibitors [34]. Another advantage of etoricoxib is the duration of action of 24 h; thus, only one dose per day is required and may increase compliance. As 90 mg and 120 mg were similarly effective (and superior to 60 mg) in patients after dental surgery [37], after knee joint replacement [36 ] and after abdominal hysterectomy [38], 90 mg once a day seems to be an appropriate perioperative dose. Improvement of pain intensity with etoricoxib after knee replacement was associated with a significant and clinical relevant reduction in opioid requirements up to 55% versus placebo within the first 7 days [36 ]. Interestingly, this opioid-sparing effect of etoricoxib increased over time after surgery and was associated with a time dependent effect on pain with knee flexion, indicating that longer duration treatment with a NSAID is improving recovery after such operations. In these outcomes, 90 mg etoricoxib was not inferior to 1800 mg ibuprofen a day, confirming the assumption of similar efficacy of nonselective NSAIDs and
0952-7907 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
&
&
&
www.co-anesthesiology.com
515
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Pain medicine
selective COX-2 inhibitors. This was also confirmed in a recently updated Cochrane Database Systematic Review on single dose oral celecoxib for acute postoperative pain in adults, which found 400 mg celecoxib to be similarly effective to 400 mg ibuprofen with median time to rescue medication of 8.4 h with 400 mg celecoxib [33]. There is some preclinical indication that NSAIDs provide some effects beyond analgesia. For example, recent animal studies suggest that systemic administration of NSAIDs prevent the development of surgery-associated memory deficits in rats [39]. However, clinical relevance of this effect needs to be confirmed. One disadvantage of most NSAIDs is the nonavailability of IV formulations of the drugs. The alternative is to use parecoxib IV (40 mg every 12 h), a selective COX-2 inhibitor for IV administration with an NNT of 1.8 for 40 mg comparable to etoricoxib [40]. Thus, parecoxib can in patients without renal risk factors replace IV paracetamol because of its greater efficacy. An IV formulation of the NSAID tenoxicam, which has been used extensively in the past, showed in combination with IV PCA after spinal fusion no improved postoperative pain control, but opioid sparing and suppression of local inflammation at the surgical site [41].
Side-effects and risk of NSAIDs One reason why NSAIDs are (in some countries or hospitals) not used frequently in the perioperative period (although they are more effective than paracetamol) is concerns about side-effects. The development of COX-2 selective drugs achieved a much better safety profile with regard to gastrointestinal ulceration, even in the acute setting, lack of platelet inhibition with reduced perioperative blood loss and no induction of bronchospasm in patients with aspirin-exacerbated respiratory disease [42]. However, most data on side-effects are coming from RCTs or data resources of patients with chronic pain and, therefore, long-term use. Safety of the perioperative use of NSAIDs and side-effects associated with short-term use in the perioperative period is currently less well studied. We will review here the current literature on side-effects and risks of NSAIDs in general and – when available – for short-term use.
Cardiovascular effects of NSAIDs It is now well known that cardiovascular events are a side-effect of all NSAIDs, nonselective ones as well as COX-2 selective ones [43]. The underlying mechanisms seem to be a permanent COX-2 inhibition 516
www.co-anesthesiology.com
rather than an imbalance between cyclooxygenase-1 (COX-1) and COX-2 as proposed earlier days. However, the mechanisms by which NSAID treatment causes cardiovascular events seem to be more multifactorial and, in part, independent of type of COX inhibition and heterogeneous between individual agents with no general class effect of the selective COX-2 inhibitors per se [44]. The largest meta-analysis on side-effects of NSAIDs has been published recently [45 ]. This meta-analysis of prospective randomized controlled trial (RCT) data characterized and quantified the cardiovascular risk of coxibs and nonselective NSAIDs. Only naproxen, a more COX-1 than COX-2 inhibitor, seems not to increase the cardiovascular risk. For short-term use, there are no new data available. The largest meta-analysis on this issue indicated that short-term use (approximately 7–10 days) of parecoxib and valdecoxib after surgery does not increase the cardiovascular risk in noncardiac surgical patients [46]. Only one large but very appealing cohort study investigated the duration of NSAID treatment and cardiovascular risk in patients with prior myocardial infarction [47]. It demonstrated that the administration of diclofenac even after short-term use (
Nonopioid analgesics for postoperative pain management Esther Pogatzki-Zahn a, Chandani Chandrasena b, and Stephan A. Schug b,c
Purpose of review Postoperative pain remains poorly treated in many patients. One reason is the inadequate use of nonopioid analgesics. This review examines the most recent findings on nonopioid analgesics and how these translate into clinical practice. Recent findings Commonly used nonopioid analgesics are paracetamol, NSAIDs and metamizol (dipyrone). For paracetamol, the efficacy is obviously inferior to NSAIDs and the risk of adverse events might have been underestimated in the past. For NSAIDs, there are increasing data supporting similar efficacy of nonselective and cyclooxygenase-2 selective NSAIDs, although the adverse effects of the latter might make them the preferred option for short-term use in the perioperative setting. Metamizol (dipyrone) is a very effective nonopioid analgesic not worldwide available; the discussion on the relevance of the very rare adverse event agranulocytosis continues. Summary Nonopioid analgesics are important components of multimodal postoperative analgesia. The selection of the most appropriate compound for an individual patient can be based more and more on ever increasing data on these important analgesics. Keywords cyclooxygenase inhibitor, dipyrone, multimodal analgesia, NSAIDs, paracetamol, postoperative pain
INTRODUCTION Postoperative pain is still poorly managed mainly because therapeutic options are not used appropriately. In fact, more minor surgical procedures seem to be associated with considerable pain indicating that evidence-based treatment recommendations might not be used in routine clinical practice [1 ]. In parallel, there has been an increasing emphasis on providing good postoperative analgesia not only for humane reasons but also to improve postoperative morbidity, possibly mortality, return to function, improved rehabilitation and earlier discharge [2]. In this context, recent studies show an impact of postoperative pain on cognitive function of aging animals, including the development of memory deficits [3]. Thus, prevention of postoperative cognitive deficits in elderly patients might be another result of improved postoperative pain management. Multimodal analgesia, that is combining drugs with different mechanism of action [such as opioids together with nonopioid analgesics (NOPAs)] [4 ], is recommended as a best practice in almost every guideline on postoperative pain [5]. However, it is &
&
still a matter of debate which nonopioid analgesic is best suited for the perioperative period. On the one hand, efficacy is very different between some NOPAs. On the other hand, like other medications, NOPAs are not without side-effects and there is increasing awareness of their potential harm. Balancing efficacy against side-effects should determine the choice of the ‘right’ analgesic in each individual patient. The following review illustrates and discusses the increasing evidence of efficacy, side-effects,
a Clinic of Anaesthesiology, Intensive Care Medicine and Pain Therapy, Muenster, Germany, bDepartment of Anaesthesia and Pain Medicine, Royal Perth Hospital and cSchool of Medicine and Pharmacology, University of Western Australia, Perth, Australia
Correspondence to Esther Pogatzki-Zahn, MD, Department of Anesthesiology, Intensive Care Medicine and Pain Therapy, University Hospital Muenster, Albert-Schweitzer-Campus 1, A1, 48149 Muenster, Germany. Tel: +49 251 8347261; fax: +49 251 88704; e-mail: [email protected] Curr Opin Anesthesiol 2014, 27:513–519 DOI:10.1097/ACO.0000000000000113
0952-7907 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
www.co-anesthesiology.com
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Pain medicine
KEY POINTS Nonopioid analgesics are an important component of multimodal perioperative analgesia. Paracetamol is a weak nonopioid analgesic with a risk of potential adverse events that might have been underestimated in the past. Nonselective NSAIDs and selective COX-2 inhibitors are similarly effective in reducing postoperative pain, have an opioid-sparing effect and reduce opioid-related side-effects; the most relevant side-effects of nonselective NSAIDs are related to the gastrointestinal tract and are reduced if selective COX-2 inhibitors are used. Cardiovascular risk factors are associated with both nonselective NSAIDs and selective COX-2 inhibitors in patients at risk. Patients with low cardiovascular risk can be safely treated perioperatively with nonselective NSAIDs as well as COX-2 inhibitors (except for cardiac surgery, in which all NSAIDs are contraindicated). Dipyrone (metamizol) has similar efficacy as NSAIDs, but a better side-effect and risk profile. Most common side-effects are reduction of blood pressure and allergic reaction, but the incidence of the feared drug-induced agranulocytosis is very rare. Therefore, dipyrone remains a commonly used nonopioid analgesic for postoperative pain management in many countries including Germany, where it is most commonly used.
indications and contraindications of nonopioids in the postoperative setting by analysis of the most recent literature.
PARACETAMOL (ACETAMINOPHEN) Paracetamol is frequently used in the perioperative period. Although the mechanism of action remains unclear, it is becoming increasingly obvious that this is a central one [6 ]. &
Efficacy of paracetamol The debate on the value and use of paracetamol as a component of multimodal analgesia continues; reports on the efficacy of paracetamol on pain after surgery are variable and sometimes rather disappointing. A recent large and very comprehensive meta-analysis [7] clearly points toward a weak effect of paracetamol compared to NSAIDs with a numberneeded-to-treat (NNT) of 3.5 when 1000 mg are used and lower paracetamol doses being less effective (NNT 4.3 with 600/650 mg). In accordance, the opioid-sparing effect of paracetamol is variable and with only very little or no effect on opioid-related side-effects [8]. In a recent systematic review in 514
www.co-anesthesiology.com
orthopedic surgery [9], six out of nine trials found better analgesia with paracetamol compared with placebo whereas three found no benefit here. A recent systematic review in children found less positive studies with paracetamol than with NSAIDs [10 ]; this was influenced by dose and route of administration. Comparative reviews against ketorolac or ibuprofen in children (and adults) indicate comparable [11] or inferior [12] analgesic potency. &&
Routes of administration for paracetamol As rectal application has an unpredictable bioavailability and should be avoided, intravenous (IV) paracetamol might be an alternative option if the oral route is not available [13]. Overall efficacy of IV paracetamol seems to be similar or – if the duration of the effect is taken into account – even less compared with that of oral paracetamol [14]. As shown by this meta-analysis, the NNT for IV paracetamol and IV propacetamol (a prodrug of paracetamol) is 4.0 at 4 h and decreases to 5.3 at 6.0 h; similarly, the authors found a 30% reduction in opioid requirements compared with placebo over the first 4 h after IV paracetamol and only a 16% reduction over the 6 h after IV paracetamol [14]. In view of the significant costs compared with oral paracetamol and potential side-effects of IV infusion including a need for venous access, IV paracetamol should only be used in patients who are not able to take oral paracetamol and only if more effective NOPAs are not indicated because of patient-related risk factors; further studies to evaluate the utility [15] and the cost-effectiveness [16] of this preparation are needed. This discussion might be influenced in favour of IV paracetamol by a meta-analysis, which showed significant reduction of postoperative nausea and vomiting by prophylactically administered IV paracetamol [17 ]. The effect was best with preoperative prophylactic administration [relative risk 0.54 (0.40–0.74) with less benefit when used before arrival in the recovery room 0.67 (0.55–0.83)] and none when used after onset of pain [1.12 (0.85– 1.48)]. It is of note that a reduction of postoperative nausea and vomiting (PONV) correlated significantly with the reduction of pain and not with reduced opioid use related to the paracetamol application [17 ]. The authors suggest either a direct link between analgesic effect and PONV reduction or a direct effect of paracetamol on PONV. The combination of paracetamol and NSAIDs has previously been shown to be superior in efficacy to paracetamol alone, but not NSAID alone [18]. A recent Cochrane Review [19] now indicates specifically for paracetamol and ibuprofen better analgesia and reduced adverse effects with the combined use &&
&&
Volume 27 Number 5 October 2014
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Nonopioid analgesics for pain management Pogatzki-Zahn et al.
compared to ibuprofen alone. However, results were based on two studies from the same study group and need further confirmation.
Side-effects and risk of paracetamol One reason for the widespread use of paracetamol worldwide is the belief that paracetamol is a harmless substance with rare systemic side-effects. This belief is increasingly challenged by more recent findings [20,21]. Ongoing discussions focus on hepatotoxicity not only because of overdose, but even with therapeutic doses [22]. Liver failure due to paracetamol intake is the most common cause of acute liver failure; the increasing number of unintentional liver failure is worrying, including nonoverdose paracetamol toxicity with a three times higher rate than all NSAIDs pooled [23]. Furthermore, the therapeutic window is narrow in adults and even less in children, and as conditions of metabolism can be altered by various factors, it seems almost impossible to predict an enhanced risk in individual patients [24 ]. The association between paracetamol intake and an increased risk of allergic asthma, rhinoconjunctivitis and eczema during the intrauterine period, infancy and adult life is becoming more obvious. Although a causal relationship between paracetamol and asthma in children after exposure in pregnancy or use in early infancy has not been established yet, there is now overwhelming epidemiological data supporting this association [25]. Although there remains a potential influence of unknown confounding factors, the findings are not affected by identified potential confounding factors. The authors of the above review call the evidence ‘compelling’ and demand randomized controlled trials and establishment of causation. This is supported by other recent literature including a review, which even asks if pediatric use should be banned [26]. This study offers also a review of the biological plausibility of an effect of paracetamol on airway reactivity via increased oxidant-induced airway inflammation due to glutathione effects. This is confirmed by the finding that the association of the risk is increased in children genetically susceptible due to antioxidant genes [27 ]. Antibiotics might add to this effect as suggested by other authors [28]. The increased risk of asthma is even linked to intrauterine exposure to paracetamol by intake of the mother [29 ]. More recently, it has been suggested that paracetamol is associated with an increased risk of hyperkinetic disorders, autisms and kryptorchismus in children whose mothers have taken paracetamol during pregnancy [30,31]. &
&&
&
Cardiovascular effects of paracetamol A potential cyclooxygenase-2 (COX-2) inhibitory effect of paracetamol suggests possible cardiovascular side-effects of paracetamol similar to those associated with NSAIDs (see below), at least in patients with risk factors [21]. However, the data on blood pressure increase in patients after exposure remain unclear; a recent systematic review confirmed this uncertainty [32]. The clinical relevance of this needs to be determined, but it is possibly not relevant for short-term use in the postoperative setting.
NSAIDS The effect of NSAIDs is related to the inhibition of cyclooxygenases.
Efficacy of NSAIDs (nonselective and cyclooxygenase-2-specific NSAIDs) There is no question about the efficacy of nonselective NSAIDs and selective COX-inhibitors in reducing acute postoperative pain, opioid requirements and opioid-related adverse events in the perioperative period. Compared with paracetamol, NSAIDs are more effective in reducing postoperative pain [7]. For instance, the NNTs are for ibuprofen 400 mg 2.5, naproxen 2.7 [7], celecoxib 400 mg 2.6 [33] and etoricoxib 90 mg 1.7 [34]. In addition, recent analyses confirm earlier results showing that NSAIDs are able to reduce opioid requirements in postoperative patients with the significant benefit of reduced side-effects of opioids, such as nausea, vomiting and sedation [35,36 ]. All available data suggest similar efficacy of nonselective NSAIDs and selective COX-2 inhibitors [34]. Another advantage of etoricoxib is the duration of action of 24 h; thus, only one dose per day is required and may increase compliance. As 90 mg and 120 mg were similarly effective (and superior to 60 mg) in patients after dental surgery [37], after knee joint replacement [36 ] and after abdominal hysterectomy [38], 90 mg once a day seems to be an appropriate perioperative dose. Improvement of pain intensity with etoricoxib after knee replacement was associated with a significant and clinical relevant reduction in opioid requirements up to 55% versus placebo within the first 7 days [36 ]. Interestingly, this opioid-sparing effect of etoricoxib increased over time after surgery and was associated with a time dependent effect on pain with knee flexion, indicating that longer duration treatment with a NSAID is improving recovery after such operations. In these outcomes, 90 mg etoricoxib was not inferior to 1800 mg ibuprofen a day, confirming the assumption of similar efficacy of nonselective NSAIDs and
0952-7907 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins
&
&
&
www.co-anesthesiology.com
515
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Pain medicine
selective COX-2 inhibitors. This was also confirmed in a recently updated Cochrane Database Systematic Review on single dose oral celecoxib for acute postoperative pain in adults, which found 400 mg celecoxib to be similarly effective to 400 mg ibuprofen with median time to rescue medication of 8.4 h with 400 mg celecoxib [33]. There is some preclinical indication that NSAIDs provide some effects beyond analgesia. For example, recent animal studies suggest that systemic administration of NSAIDs prevent the development of surgery-associated memory deficits in rats [39]. However, clinical relevance of this effect needs to be confirmed. One disadvantage of most NSAIDs is the nonavailability of IV formulations of the drugs. The alternative is to use parecoxib IV (40 mg every 12 h), a selective COX-2 inhibitor for IV administration with an NNT of 1.8 for 40 mg comparable to etoricoxib [40]. Thus, parecoxib can in patients without renal risk factors replace IV paracetamol because of its greater efficacy. An IV formulation of the NSAID tenoxicam, which has been used extensively in the past, showed in combination with IV PCA after spinal fusion no improved postoperative pain control, but opioid sparing and suppression of local inflammation at the surgical site [41].
Side-effects and risk of NSAIDs One reason why NSAIDs are (in some countries or hospitals) not used frequently in the perioperative period (although they are more effective than paracetamol) is concerns about side-effects. The development of COX-2 selective drugs achieved a much better safety profile with regard to gastrointestinal ulceration, even in the acute setting, lack of platelet inhibition with reduced perioperative blood loss and no induction of bronchospasm in patients with aspirin-exacerbated respiratory disease [42]. However, most data on side-effects are coming from RCTs or data resources of patients with chronic pain and, therefore, long-term use. Safety of the perioperative use of NSAIDs and side-effects associated with short-term use in the perioperative period is currently less well studied. We will review here the current literature on side-effects and risks of NSAIDs in general and – when available – for short-term use.
Cardiovascular effects of NSAIDs It is now well known that cardiovascular events are a side-effect of all NSAIDs, nonselective ones as well as COX-2 selective ones [43]. The underlying mechanisms seem to be a permanent COX-2 inhibition 516
www.co-anesthesiology.com
rather than an imbalance between cyclooxygenase-1 (COX-1) and COX-2 as proposed earlier days. However, the mechanisms by which NSAID treatment causes cardiovascular events seem to be more multifactorial and, in part, independent of type of COX inhibition and heterogeneous between individual agents with no general class effect of the selective COX-2 inhibitors per se [44]. The largest meta-analysis on side-effects of NSAIDs has been published recently [45 ]. This meta-analysis of prospective randomized controlled trial (RCT) data characterized and quantified the cardiovascular risk of coxibs and nonselective NSAIDs. Only naproxen, a more COX-1 than COX-2 inhibitor, seems not to increase the cardiovascular risk. For short-term use, there are no new data available. The largest meta-analysis on this issue indicated that short-term use (approximately 7–10 days) of parecoxib and valdecoxib after surgery does not increase the cardiovascular risk in noncardiac surgical patients [46]. Only one large but very appealing cohort study investigated the duration of NSAID treatment and cardiovascular risk in patients with prior myocardial infarction [47]. It demonstrated that the administration of diclofenac even after short-term use (
