Nonsteroidal Anti-Inflammatory Drugs for Postoperative Pain A Focus on Children

Transcription

1 REVIEW ARTICLE Pediatr Drugs 2003; 5 (2): /03/ /$30.00/0 Adis International Limited. All rights reserved. Nonsteroidal Anti-Inflammatory Drugs for Postoperative Pain A Focus on Children Hannu Kokki Department of Anaesthesiology and Intensive Care, Kuopio University Hospital, Kuopio, Finland Contents Abstract Mechanism of Action and Analgesic Effect Pharmacokinetics Administration Routes Types of Analgesia Pre-Emptive Analgesia Proactive Pain Management Multimodal Analgesia Combining Nonsteroidal Anti-Inflammatory Drugs (NSAIDs) with Acetaminophen (Paracetamol) Use of NSAIDs During Surgery: Analgesic Efficacy and Tolerability Day-Case Surgery Myringotomy Adenoidectomy Tonsillectomy Dental Surgery Strabismus Herniotomy Urologic Surgery Orthopedic Surgery General Pediatric Surgery Summary Special Patient Groups Newborns Neurosurgery Children with Asthma Critically Ill Longer Term Use of NSAIDs Contraindications and Adverse Effects Effect on Renal Function Hematologic Effects Bone and Ligament Healing Gastric Mucosal Damage Fever Reye s Syndrome Parent Guidance Conclusions Abstract Pain is a common symptom after surgery in children, and the need for effective pain management is obvious. For example, after myringotomy, despite the brief nature of the procedure, at least one-half of children have significant pain. After more extended surgery, such as tonsillectomy, almost all children have considerable pain longer than 7 days. Nonsteroidal anti-inflammatory drugs (NSAIDs) are useful for postoperative pain management because

2 104 Kokki surgery causes both pain and inflammation. Several pediatric studies indicate NSAIDs are effective analgesics in the management of mild and moderate pain. In the treatment of severe pain, NSAIDs should be given with acetaminophen (paracetamol) or opioids, and the use of an appropriate regional analgesic technique should be considered. NSAIDs are more effective in preventing pain than in the relief of established pain. Pain following surgery is best managed by providing medication on a regular basis, preventing the pain from recurring. This proactive approach should be implemented for any procedure where postoperative pain is the likely outcome. In children, the choice of formulation can be more important than the choice of drug. Intravenous administration is preferred for children with an intravenous line in place; thereafter mixtures and small tablets are feasible options. Children dislike suppositories, and intramuscular administration should not be used in nonsedated children. Ibuprofen, diclofenac, ketoprofen and ketorolac are the most extensively evaluated NSAIDs in children. Only a few trials have compared different NSAIDs, but no major differences in the analgesic action are expected when appropriate doses of each drug are used. Whether NSAIDs differ in the incidence and severity of adverse effects is open to discussion. Because NSAIDs prevent platelet aggregation they may increase bleeding. A few studies indicate that ketorolac may increase bleeding more so than other NSAIDs, but the evidence is conflicting. Severe adverse effects of NSAIDs in children are very rare, but it is important to know about adverse effects in order to recognize and treat them when they do occur. NSAIDs are contraindicated in patients in whom sensitivity reactions are precipitated by aspirin (acetylsalicylic acid) or other NSAIDs. They should be used with caution in children with liver dysfunction, impaired renal function, hypovolemia or hypotension, coagulation disorders, thrombocytopenia, or active bleeding from any cause. In contrast, it seems that most children with mild asthma may use NSAIDs. Management of pain in children is changing rapidly as a result of improvements in the appreciation of pediatric pain and pharmacological knowledge, and the availability of appropriate drug formulations. Pain is also a common symptom after surgery in children. For ethical and humanitarian reasons pain should be effectively controlled in all age groups. Postoperative pain treatment in pediatric patients should be improved because it has previously been a popular misconception that children do not feel pain as severely as adults do, and that the magnitude and duration of its impact may be less than in adults. [1,2] An increasing number of publications indicate that children in hospital receive less medication following the same type of procedure compared with adults. [3-5] Nonsteroidal anti-inflammatory drugs (NSAIDs) are useful for postoperative pain management because surgery can cause both pain and inflammation. NSAIDs are now increasingly used for postoperative analgesia in children. [5,6] Drugs used for humans should be extensively tested to ensure their efficacy and tolerability before preparations are approved for clinical use. This is not always the case for medicines used in children. It is estimated that up to 90% of medical products used in children have not been specifically evaluated for use in children. [7] On the contrary, in some other therapeutic areas several NSAIDs have been evaluated in children. However, many NSAIDs commonly used have not been authorized for pediatric use, and off-label use of NSAIDs in children seems to be common. [8] Specific studies in children are necessary because there are age-related differences in the drug handling and effects that may necessitate dose adjustments to those commonly used in adults. Small children may also need different drug formulations compared with those used in adults; for example, swallowing capsules is often impossible for small children and other dosage forms should be used instead. This is important because, for example, in Western Europe one-fifth of the total population are children, and in some other parts of the world children represent one-half of the population. In the European Union there are 75 million children aged between 0 and 16 years, and 5 10% of them undergo surgery annually. [9] Hence, the need for effective postoperative pain treatment is obvious. The aim of this review is to describe the present findings on the use of NSAIDs in the management of postoperative pain in children. In addition, suggestions on their tolerability and effective use are given. 1. Mechanism of Action and Analgesic Effect Most NSAIDs exhibit a spectrum of analgesic, anti-inflammatory, antiplatelet and antipyretic actions. In general, NSAIDs are useful for postoperative pain management because surgery can cause both pain and inflammation. The major analgesic effect of NSAIDs appears to result from the inhibition of prostaglandin synthesis. [10] Studies suggest that the analgesic activity of NSAIDs principally results from a peri-

3 NSAIDs for Postoperative Pain 105 pheral action. In addition, the anti-inflammatory effect of NSAIDs may contribute to their analgesic effect, but NSAIDs do not appear to affect opioid receptors. [11] Other proposed antinociceptive mechanisms for NSAIDs include decreased production of leukotrienes, activation of serotonin pathways and inhibition of excitatory amino acids, and N-methyl-D-aspartate receptormediated hyperalgesia. [12] NSAIDs exert their analgesic effects mainly by inhibiting prostaglandin synthesis at the level of cyclo-oxygenase (COX), thereby blocking the production of prostaglandins that stimulate nociceptors in the peripheral nervous system. Moreover, some NSAIDs also seem to have a central analgesic action, and are proposed to inhibit central prostaglandin biosynthesis. [13,14] Some NSAIDs are rapidly and readily distributed into the central nervous system, with cerebrospinal fluid concentrations of free drug approximately equivalent to those in plasma. [15] The action of NSAIDs was previously believed to result from the inhibition of a single COX enzyme system. Subsequently, the existence of two separate COX isoenzymes, namely COX-1 and COX-2, has been established. [16] COX-1 is a constitutive enzyme that is expressed in most tissues, blood monocytes and platelets; COX-1 is involved in thrombogenesis (e.g. promotion of platelet aggregation), maintenance of the gastric mucosal barrier, and renal function (e.g. maintenance of renal perfusion). [11] COX-2 is an inducible enzyme that is principally found at the site of inflammation, although it is also expressed constitutively in the brain, kidney and reproductive organs. COX-2 is expressed within 2 12 hours in response to cytokines and growth factors. At clinically relevant concentrations, NSAIDs inhibit COX-2 in a slow, time-dependent manner. [11] Whether inhibition of both COX isoenzymes is necessary for achieving a maximal analgesic effect with these drugs is still open to discussion. Because NSAIDs have a rapid analgesic action in the management of postoperative pain, it is also thought that inhibition of COX-1 or other mechanisms are involved in producing analgesia with NSAIDs. [12,14] 2. Pharmacokinetics There are some differences in the responses to NSAIDs, particularly in younger children, compared with adults. The pharmacodynamic differences are not fully elucidated and, in general, our knowledge of infants <6 months of age is sparse. Altered pharmacokinetics may explain the difference in responses, rather than actual altered responsiveness. Moreover, in pre-verbal children it is often difficult to recognize all effects in responses. The pharmacokinetics of NSAIDs during surgery have been evaluated in some studies. The absorption of drugs may be delayed during the immediate postoperative period. These suggestions are supported by the findings with acetaminophen (paracetamol). The absorption of acetaminophen administered 1 hour before induction is rapid, with a mean half-life for absorption of 4.5 minutes without a detectable lag time; however, on the first morning after tonsillectomy, absorption is much slower, i.e. absorption half-life of 44 minutes. [17,18] Delayed gastric emptying and the drug formulation may affect the usefulness of some oral dosage forms in postoperative pain management. In a study where enteric-coated diclofenac tablets were given in the first postoperative morning, a mean absorption lag time of 1 hour, and time to reach maximum concentration (t max ) of 2 hours after administration were found. Therefore, the enteric-coated tablets are considered unsuitable for the initiation of pain management after surgery. [18] With both a continuous infusion and regular bolus administration, it takes five half-lives for a drug to achieve a steady blood concentration. Therefore, it is common practice to start the treatment with a bolus dose, followed by a continuous infusion to keep a desired blood concentration. [19] In general, drug absorption by mouth and parenteral routes is similar in children and adults. In a study in the USA, [20] bodyweight-normalized pharmacokinetics of intravenous ketorolac did not differ among younger children and adolescents, and were similar to those reported for adults. Hamunen et al. [21] evaluated stereoselective pharmacokinetics of racemic ketorolac in children, adolescents and adults, and it was shown that the clearance, volume of distribution (Vd), and elimination half-life were higher for the pharmacological active (S)-ketorolac than for the less active (R)-enantiomer. The Vd of the active (S)-enantiomer was higher in children than adolescents and adults, but the values for clearance were similar. Because of the higher Vd, the elimination half-life was also higher in children than adults. On the contrary, the pharmacokinetics of the (R)-enantiomer were essentially unaffected by age. Trials have also shown no major differences in the pharmacokinetics of ketoprofen between healthy adults and children. For example, the elimination half-life of ketoprofen (around 2 hours) is similar to that reported in adults. [22] In addition, as shown in adults, no evidence of accumulation of ketoprofen was noticed after 24 hours continuous infusion in children. [19] Studies have revealed that the rate and extent of absorption of, for example, ketoprofen is comparable after intramuscular and oral administration, syrup and tablets, and therefore, there is no justification for using intramuscular or rectal administration in awake children. [23]

4 106 Kokki Most NSAIDs undergo hepatic metabolism by phase I and II biotransformation enzymes. Conjugation occurs mainly with glucuronic acid and sulfate. About two-thirds of the parent drug or its metabolites are excreted in the urine, and the remaining one-third are excreted in the feces. [12] 3. Administration Routes For pain treatment in children, the choice of formulation can be more important than the choice of drug. NSAIDs are available in several pharmaceutical dosage forms, and hence different routes of administration may be used. However, only some of the preparations are approved for children. There are some limiting factors to all different routes of administration. The intravenous route is accurate and practical during the perioperative period as long as the patient has an intravenous line in place. [24-27] Rapid intravenous bolus of NSAIDs may cause bradycardia, [28] therefore NSAIDs should be given by a slow injection, i.e. >5 minutes. Intramuscular administration is effective, but because of the injection pain it should be avoided in nonsedated children. [29] NSAID administration by mouth is supported by many because the oral route is cheap, pleasant and effective, and the medications can be administered as small tablets or as a mixture. [27,30,31] However, the oral route cannot be used when a patient is unable to swallow, is unconscious, nauseated, or is not co-operative. The intravenous or rectal route can be useful in these children. [32-35] In a UK study, [36] parents opinions about the acceptability of different routes of administration of analgesia for their children were evaluated. The oral route was considered to be the most convenient by most parents, followed by intramuscular and intravenous routes. The majority of parents considered the rectal route to be the most unpleasant way of giving analgesics. The parents opinion is supported by other studies showing that children dislike suppositories. [37] Few trials have compared different administration routes of NSAIDs in children. Diclofenac by mouth seems to perform better than the same dosage administered rectally in children undergoing tonsillectomy. [38] Different administration routes of ketoprofen have been studied extensively in children. In general there does not seem to be any clinically significant difference in analgesic effect, or in the rate and extent of absorption and the elimination of ketoprofen between different dosages forms. [39,40] Therefore, there is no justification from the pharmacologic point of view to use intramuscular or rectal administration of ketoprofen in awake children. NSAIDs should preferably be given intravenously for children with an intravenous line in place, and thereafter mixtures and small tablets are also feasible options for young children. In general, rectal and intramuscular administration should be avoided in nonsedated children. However, in children without an intravenous line who are vomiting, rectal administration can be used. 4. Types of Analgesia 4.1 Pre-Emptive Analgesia The concept of pre-emptive analgesia was first described by Wall. [41] It is based on the assumption that administration of an analgesic drug before nociceptive input can prevent sensitization and thus ameliorate postoperative analgesia. The aim of preemptive analgesia is to prevent a prolonged change in the central nervous system function by blocking of afferent input before the surgical stimulation may eliminate central sensitization, and therefore prevent amplification and prolongation of postoperative pain. [42] However, the results of studies have been inconclusive. In adults, some studies report significantly better analgesia with NSAIDs administered before surgery compared with administration of the same drug after surgery, [43] while most reports show no such benefit. [44-49] The pre-emptive effect of NSAIDS has been evaluated in some studies in children. Preoperative or postoperative treatment with ketorolac during tonsillectomy seems to be equally effective. Römsing et al. [50] evaluated the analgesic efficacy of ketorolac intravenously, given before or after surgery in children undergoing tonsillectomy. Rescue analgesic consumption was less in the immediate postoperative period for the children receiving preoperative ketorolac, but no differences were found between the two groups in the first 24 hours after surgery. During appendectomy no pre-emptive effect of analgesic treatment with diclofenac was found, because the duration of rescue analgesia treatment was similar across the study groups. [51] The pre-emptive effect of ketoprofen has been evaluated extensively in children in postoperative pain treatment after throat surgery. However, in three large studies no significant differences have been shown in pain intensity, proportion of children needing rescue analgesics, number of analgesic doses, or duration of pain at home between patients who received the first dose of ketoprofen before adenoidectomy or tonsillectomy and those who received the first dose at discharge. [40,52-54] In conclusion, the timing of NSAID administration is considered to be important for immediate postoperative pain, but studies do not support the existence of a pre-emptive analgesic effect.

5 NSAIDs for Postoperative Pain Proactive Pain Management NSAIDs are expected to be more effective in preventing pain than in the relief of established pain. [55] With NSAIDs it can take 1 2 hours before the maximal analgesic effect is seen. The concept of proactive pain management with non-opioid analgesics is best achieved when the drug is administered as early as possible, before the pain has broken through, and continued on a regular basis for as long as the pain is expected to last. [55] Using this approach to pain management, attempts are not made to reduce the amount of pain medication. However, postoperative pain is better controlled when analgesics are given on a regular basis and, therefore, unwanted breakthrough of pain and unnecessary suffering is avoided. Some studies support these expectations, and clinical experience also shows that proactive pain treatment with NSAIDs is beneficial for known painful surgery. [56,57] Studies have detected a substantial incidence of problematic behavioral changes in children after surgery, with pain being a significant factor predicting these problematic changes. [58,59] For example, after tonsillectomy, poor fluid intake, sleep disturbance, behavioral changes and enuresis have been described. [54,59] The use of NSAIDs may also be beneficial from a psychological point of view. The concept of beneficial effects of proactive treatment is supported by a recent study of day-case adenoidectomy, in which proactive pain treatment with ketoprofen resulted in a negligible incidence of behavior troubles. [40] 4.3 Multimodal Analgesia Multimodal analgesia consists of a combination of analgesic regimens to provide more effective analgesia while reducing the incidence and severity of adverse events. [42] Combining opioids and local anesthetics with NSAIDs, with or without acetaminophen, to produce analgesia has been found particularly useful in pediatric surgery, and has assumed increasing importance in the management of postoperative pain. [32,51,60,61] NSAIDs are commonly used as part of multimodal analgesia because they decrease the postoperative opioid requirements, thereby decreasing the severity of opioid-related adverse effects. [5] Moreover, NSAIDs may enhance the analgesia achieved by the analgesic regimen, thus providing improved patient satisfaction. During appendectomy, multimodal analgesia with rectal diclofenac, wound infiltration with bupivacaine, and intravenous morphine seemed to perform satisfactorily. Pain scores were low, both at rest and on movement, and the need for rescue analgesic was significantly less than in children treated with wound infiltration and intravenous morphine alone. [51] In children who have undergone major surgery, several studies have confirmed that the addition of an NSAID to an epidural regimen is well tolerated and efficacious, and may reduce opioidassociated adverse effects. [62-64] Combining Nonsteroidal Anti-Inflammatory Drugs (NSAIDs) with Acetaminophen (Paracetamol) Mild pain may usually be controlled with NSAIDs alone, but in more severe pain a combination of analgesics should be used. In patients with moderate pain, concurrent use of two NSAIDs is not recommended, but pain relief is assumed to be improved by the addition of acetaminophen to an NSAID. With the recent launch of an intravenous preparation of acetaminophen (approved for children >33kg), this agent is also considered to be a valuable drug in perioperative pain management in children. Some trials have shown that the combination of an NSAID with acetaminophen performs significantly better than acetaminophen alone. [50,65] However, drug combinations with the various NSAIDs may perform differently. Pickering et al. [66] have shown that the addition of ibuprofen, but not rofecoxib, to acetaminophen reduced the need for early analgesia by 50% in children undergoing tonsillectomy. Combining NSAIDs with acetaminophen did not increase the incidence of adverse effects or bleeding. On the contrary, whether the combination of an NSAID with acetaminophen performs better than an NSAID alone is sparsely demonstrated, but there are some papers that describe the concurrent use of NSAIDs with acetaminophen in children. Two studies describe the use of diclofenac with acetaminophen in children undergoing adenoidectomy and tonsillectomy. The combination seems to be highly effective because only 10 out of 110 children receiving diclofenac with acetaminophen in these two studies required any rescue analgesia. [30,38] A combination of an NSAID with acetaminophen has been shown to be associated with a lower incidence of postoperative nausea and vomiting (PONV) than an NSAID with opioid combination. [67] During strabismus surgery, diclofenac combined with acetaminophen provided a significant analgesic effect, with a low incidence of PONV (12%). [68] Church [69] described an analgesic technique with regular administration of ibuprofen and acetaminophen by mouth for 7 days in children with tonsillectomy. He reported that the drug regimen performed well, better than the previous protocol which had no plan for pain management. However, it should be noted that this was a cases series survey and therefore no firm conclusion can be made. The type of surgery may influence the rationale to use a combination of an NSAID with acetaminophen. During appendectomy, the analgesic effect of wound infiltration with bupivacaine was significantly improved by concurrent use of rectal diclofenac,

6 108 Kokki but no additive effect of acetaminophen was found. Moreover, the extent of significant pain was comparable in the diclofenac plus acetaminophen and the diclofenac monotherapy groups. [51] Further studies are awaited to see whether the analgesia achieved with the combination of an NSAID plus acetaminophen is better than with either drug alone used in appropriate dosages. 5. Use of NSAIDs During Surgery: Analgesic Efficacy and Tolerability Children frequently undergo different types of surgery and healthcare professionals should have a plan for postoperative pain treatment. NSAIDs are considered to be more effective in preventing pain than in the relief of established pain in children. [70] The analgesic efficacy and tolerability of different NSAIDs in treating postoperative pain in children has been evaluated in several randomized, controlled trials. However, several case series and retrospective studies on NSAIDs in children have also been reported and it should be noted that they do not provide a high level of evidence of tolerability or efficacy of NSAIDS in children. Pain assessment in young children is difficult; therefore, analgesic studies in infants and younger children are challenging. In most trials, the analgesic efficacy of NSAIDs has been evaluated by comparing observed pain scores. This may limit the interpretation of results compared with trials in older children and in adults who are able to self-express the level of pain and possible adverse effects. Most of the trials are placebo-controlled, or compare the analgesic efficacy of NSAIDs with acetaminophen or opioids. Only a limited number of trials compared different NSAIDs for postoperative pain treatment in children. However, with an understanding of the main mechanism of the analgesic action of NSAIDs, no major differences would be expected to occur when appropriate doses of each drug are used. Whether the difference between NSAIDs on platelet aggregation has any clinical significance is open to discussion. [71] NSAIDs have been compared in four trials. Nishina et al. [72] compared diclofenac with flurbiprofen in children undergoing strabismus surgery, but no significant differences were observed between the two NSAIDs in the pain scores or PONV. Diclofenac has been compared with ibuprofen in children undergoing orthopedic surgery, but no differences were found between the two NSAIDs. [73] Compared with rectal diclofenac, tenoxicam-induced analgesia developed more slowly, but was sustained longer during tonsillectomy. No differences between the groups were noted in the incidence or severity of adverse effects. [74] Ibuprofen and rofecoxib as an adjunct to acetaminophen were compared during tonsillectomy. While ibuprofen halved the need for rescue analgesia, rofecoxib failed to improve analgesia compared with acetaminophen alone. [66] NSAIDs are expected to have a ceiling on their analgesic effectiveness; however, only a few studies have compared different doses of NSAIDs. Maunuksela et al. [75] compared ketorolac with morphine in a dose titration study and found that a dose of 0.2 mg/kg ketorolac did not provide sufficient analgesia, but a dose of 0.5 mg/kg was as effective as morphine 0.1 mg/kg. Purday et al. [76] compared ketorolac at doses of 0.75, 1.0 and 1.5 mg/kg, but did not find that analgesic efficacy was dose-dependent. Mikawa et al. [77] has shown that intravenous flurbiprofen 1 mg/kg provides a significantly better analgesic efficacy than a dose of 0.5 mg/kg. Ketoprofen has already been shown to have a significant analgesic efficacy at a dose of 0.3 mg/kg, [29] while with higher doses (up to 3 mg/kg), enhanced analgesia is achieved without any increase in adverse effects. [78,79] Suggested doses of some NSAIDs for postoperative pain management in children over 3 months of age are given in table I. 5.1 Day-Case Surgery In recent years there has been a trend towards performing more surgery on a day-case basis, and among children day-case surgery comprises over 60% of the total surgical case load. Chil- Table I. Suggested dosages of some NSAIDs for postoperative pain management in children >3 months of age a Agent Single doses (mg/kg) Frequency (hourly) Maximal daily dose (mg/kg) Comment References Diclofenac Also IV preparation 18,38,51,68,72,74,80-90 Ibuprofen ,91-96 Flurbiprofen Also IV preparation 72,77 Ketoprofen Also IV preparation 15,19,22,23,26,31,34,35,53,54,57,62-64,78,79,97,98 Ketorolac Also IV preparation 25,50,75,76, a The same doses may be used intravenously, by mouth, and rectally. IV = intravenous.

7 NSAIDs for Postoperative Pain 109 dren make excellent candidates for day surgery as they are usually healthy, free of systemic disease, and typically require straightforward, minor, or intermediate surgical procedures. [117] NSAIDs are effective analgesics in the management of mild and moderate pain and, therefore, ideal for day-case surgery when pain is expected to be minimal. In the treatment of more severe pain, NSAIDs should be given with acetaminophen and/or opioids. Nowadays it is widely accepted that the analgesic technique should include proactive NSAIDs, if not contraindicated, for any procedure where postoperative pain is the likely outcome. [32] Children and caregivers should be actively involved in both the assessment and management of pain after day-case surgery. They should be informed that pain is common after day-case surgery. Children may have no pain, or only mild pain, in the initial postoperative period, but recent studies indicate that after discharge, pain is frequently more severe than expected. Most children need effective analgesic treatment at home to ensure calm recovery, and the immediate postoperative comfort obtained by proactive analgesia needs to be followed by analgesics given on a continuous basis for the first days after surgery. [54,55,69,118,119] Studies suggest that the provision of preparatory information to parents creates accurate expectations and reduces pain for children. [37,120,121] In studies where all oral information has been reinforced with leaflets detailing appropriate pain management after adenoidectomy, over 90% of parents have provided pain management at home as advised. [39,40] 5.2 Myringotomy Myringotomy and ventilation tube placement is commonly performed for the treatment of acute and secretory otitis media in children. Despite the brief nature of the procedure with limited tissue trauma, at least one-half of the children had significant pain after myringotomy and placement of ventilation tubes. [122] It is obvious that these children should be provided with analgesic treatment after the procedure. Some trials have evaluated the use of NSAIDs in the prevention and treatment of pain after myringotomy, but the results are inconclusive. A study in the USA [99] showed that the children treated with ketorolac had significantly lower postoperative pain scores and required significantly less rescue analgesia than children receiving either acetaminophen or placebo during bilateral myringotomy. On the contrary, Bean-Lijewski and Stinson [100] found that acetaminophen and ketorolac provided similar, effective analgesia in children undergoing myringotomy. One trial did not find any benefit of preoperatively administered ibuprofen or acetaminophen over placebo. However, the observation period after myringotomy was only 1 hour; this may be considered too short to show any significant difference between the three treatment groups. [91] 5.3 Adenoidectomy Adenoidectomy is one of the most common surgical procedures during childhood. After adenoidectomy the recovery is often straightforward, and most children need analgesic treatment for just 2 or 3 days after discharge. However, in some children, severe pain may persist for up to 7 days and these children need special attention. [123] NSAIDs have been used with good results for postoperative analgesia after adenoidectomy. During adenoidectomy, rectal diclofenac seems to perform at least as well as acetaminophen [80] and racemic ketamine 6 mg/kg orally. [81] No increase in intraoperative bleeding or other adverse effects was noticed. Walmsley [30] describes his technique of giving diclofenac orally 30 minutes preoperatively, and two other doses on the evening of surgery and the next morning. Acetaminophen was used for rescue analgesia, and only one out of the 30 children required any perioperative opioid. However, the report by Walmsley is a case series and does not provide a high level of evidence; no definite conclusion can be drawn regarding the efficacy of diclofenac in this setting. Viitanen and Annila [124] have shown that rectal ibuprofen alone is not sufficient analgesia for children undergoing day-case adenoidectomy, but combined with tramadol it provides an excellent analgesia for most children without any increase in adverse effects. Studies with ketoprofen at doses of 0.3 to 3 mg/kg, administered either before or during surgery, perform significantly better than placebo without an increase in adverse effects such as intraor postoperative bleeding. However, as shown with ibuprofen, [124] one-half of the patients receiving ketoprofen required an opioid during surgery to ensure a calm recovery. [26,34,35,78,79,97] The use of perioperative ketoprofen does not cause clinically significant bleeding and it does not delay discharge following adenoidectomy. [125] In addition, proactive management with ketoprofen for 72 hours after surgery may improve the behavioral recovery after surgery. [40] 5.4 Tonsillectomy Particularly challenging to healthcare professionals is the provision of effective and well-tolerated pain management for children undergoing tonsillectomy. After tonsillectomy, almost all children have considerable pain, which may last longer than 7 days. [69,126] In many children, not only drinking and eating, but even speech, may be painful. Moreover, sleeping patterns are

8 110 Kokki commonly disturbed because of persistent pain. [54] Studies indicate that post-tonsillectomy pain is poorly managed; this is not only an ethical problem, but poor pain management may also lead to increased utilization of health services. [69] Tolerability of analgesic treatment is a special concern because tonsillectomy is associated with a risk of postoperative bleeding and PONV. Opioids are still frequently used during tonsillectomy, but extensive use of opioids may increase the incidence of PONV. [65] NSAIDs may increase blood loss during surgery and increase the risk of postoperative bleeding, and therefore the use of NSAIDs during tonsillectomy has been challenged. On the contrary, because NSAIDs do not induce respiratory depression, [ ] They are considered beneficial in patients who have symptoms of obstructive sleep apnea syndrome. Several NSAIDs have been used during tonsillectomy (table II). Tenoxicam 0.75 mg/kg intramuscularly was compared with morphine 0.2 mg/kg intramuscularly, [130] and tenoxicam 0.4 mg/kg intravenously with/without fentanyl was compared with rectal diclofenac 1 mg/kg with/without fentanyl. [74] Neither NSAID seemed to perform sufficiently, although the incidence of PONV was low when the agents were used as part of an analgesic regimen. The incidence of PONV was around 30% less in the children treated with tenoxicam than in the children treated with diclofenac, [74] and three times less than in the patients treated with morphine. [130] Diclofenac may offer advantages over opioids with regards to convenience of use in the treatment of pain after tonsillectomy in children. Two studies indicate that children treated with diclofenac alone are less drowsy after surgery than their counterparts who have been treated with opioids alone for pain management. [82,127] However, most children in the diclofenac-treated groups needed supplementary analgesia after surgery, and the incidence of PONV was high (50%) across the study groups. The selection of opioid may affect the severity of PONV; fentanyl combined with diclofenac plus acetaminophen has been shown to be associated with less severe PONV than morphine combined with diclofenac plus acetaminophen. The overall incidence was high in both groups (70% in the fentanyl group and 78% in the morphine group), but children who received morphine had more episodes of vomiting (median 2, range 0 7) than children who received fentanyl (median 1, range 0 3). [131] In another study comparing high-dose acetaminophen (90 mg/kg/24h), with diclofenac (2 3 mg/kg/24h), diclofenac was not more effective than acetaminophen, but the incidence of PONV was significantly less in the diclofenac-treated group than in acetaminophen recipients. [83] Diclofenac, given rectally [74] or orally, [30] seems to perform at least as well as intramuscular diclofenac. A minor increase in intraoperative blood loss with diclofenac does not seem to be clinically significant, and the incidence of bleeding necessitating a return to theatre has been relatively low in patients treated with diclofenac. [83,84] Diclofenac has been compared with tramadol [85] and acetaminophen [83] in pain treatment after discharge. The outcome across the study groups was otherwise similar, but the children in the tramadol-treated group experienced more PONV than the children in the non-opioid-treated groups. One study described the use of diclofenac intravenously for rescue analgesia during the first 6 hours after tonsillectomy, followed by regular ibuprofen/acetaminophen/tramadol management orally after discharge. The NSAID/acetaminophen/tramadol combination provided effective analgesia with a relatively low incidence of PONV (12 out of 50 children vomited). [67] In a case series study concerning 268 children with tonsillectomy, effective analgesia was achieved with rectal diclofenac during surgery, followed by ibuprofen plus acetaminophen orally four times daily for 7 days. [69] Several studies describe the tolerability and efficacy of ketorolac during tonsillectomy. Sutters et al., [101] Gunter et al., [102] Mather and Peutrell, [65] and Römsing et al. [50] have evaluated the analgesic efficacy and tolerability of ketorolac in placebo- and active counterpart-controlled trials. They demonstrated a significant analgesic effect and low incidence of PONV without any evidence of increased bleeding. On the contrary, some studies have demonstrated that ketorolac may increase bleeding during tonsillectomy. One of the initial studies [103] compared ketorolac 1 mg/kg with acetaminophen 35 mg/kg rectally in 50 children, aged between 2 and 15 years, undergoing tonsillectomy. The two drugs provided a similar analgesic effect, but both the intra- and postoperative bleeding was significantly higher in the ketorolactreated group. Most of the patients in both groups required rescue analgesia during the first 24 hours after surgery. Another study was terminated because five out of 35 patients treated with ketorolac were admitted to hospital because of bleeding; two of these patients required surgical exploration. [104] However, in a recent study in the USA, [105] ketorolac was not found to increase the incidence of post-tonsillectomy hemorrhage and, furthermore, was associated with a decrease in the length of hospital stay, as well as a decreased likelihood of overnight hospital stay after surgery. However, this was a retrospective study involving 310 patients, of whom 213 received ketorolac at the end of surgery and 97 received placebo. Therefore, any conclusions drawn should be interpreted with caution. Pickering et al. [66] compared the analgesic effectiveness of acetaminophen alone with acetaminophen combined with either ibuprofen or rofecoxib. The addition of ibuprofen, but not rofecoxib, to acetaminophen reduced the need for early analgesia by

9 Adis International Limited. All righs reserved. Pediatr Drugs 2003; 5 (2) Table II. Use of nonsteroidal anti-inflammatory drugs (NSAIDs) to improve analgesia in children undergoing tonsillectomy NSAID/control Time No. of patients Rescue analgesic Need for Pain control Bleeding Vomiting Reference rescue analgesic Ketoprofen IV/placebo IV At induction 89/20 Oxycodone /± /± /± /± 53 Tenoxicam IV/diclofenac PR At induction 63/58 Morphine/acetaminophen /± /± /± /± 74 Ibuprofen + acetaminophen (paracetamol) PO/rofecoxib + acetaminophen PO/ acetaminophen PO Preoperatively 40/40/18 Ibuprofen/codeine 17/27/13 / /± / /± / /± 66 Ketorolac IM/acetaminophen PO/morphine IV Preoperatively 28/24/28 Morphine / /± NR NR / /± 65 Diclofenac PR/papaveretum IM/placebo At induction 20/20/20 Papaveretum/acetaminophen / /± / /± / /± NR 127 Tenoxicam IM/morphine IM At induction 25/25 Morphine/acetaminophen /± /± /± /± 130 Diclofenac PR/meperidine (pethidine) IM/placebo At induction 25/25/25 Meperidine / /± / /± NR / /± 82 Diclofenac PO/acetaminophen PO First postoperative day 24/24 Fentanyl NR /± /± /± 83 Diclofenac IM/papaveretum IM At induction 91/92 Papaveretum/acetaminophen /± NR /± /± 84 Diclofenac PO/tramadol PO At end of surgery 25/24 Diclofenac/tramadol /± /± /± NR 85 Ketorolac IM/placebo IM At end of surgery 45/42 Fentanyl /± /± /± NR 101 Ketorolac IV/morphine IV At end of surgery 49/47 Morphine /± /± /± /± 102 Ketorolac IV/acetaminophen PR At induction 25/25 Morphine/acetaminophen /± /± /± NR 103 Ketorolac IV/codeine IM At induction 35/29 Morphine NR NR /± /± 104 Ketorolac IV/placebo IV At end of surgery 213/97 NR NR NR /± NR 105 Ibuprofen PO/acetaminophen PO At end of surgery 55/55 Codeine NR /± /± /± 92 IM = intramuscular; IV = intravenous; NR = not reported; PO = oral; PR = rectal; = same, similar; = less; = more; ± = reference. NSAIDS for Postoperative Pain 111

10 112 Kokki 50% compared with acetaminophen alone. In a nonblinded study in the USA, [92] ibuprofen was as effective as acetaminophen with codeine for postoperative pain control in children after tonsillectomy. The incidence of bleeding (8%) was similar in both groups, but significantly less PONV was noted in children treated with ibuprofen. In a small pilot study, involving only 27 patients, Harley and Dattolo [93] compared the efficacy and tolerability of ibuprofen with acetaminophen plus codeine for pediatric patients who had undergone tonsillectomy. The pain relief was more effective in children treated with acetaminophen plus codeine on days 1 and 3, but no differences were detected between the two groups on day 5. Children in the ibuprofen-treated group required analgesic treatment for a significantly longer period, but no differences were found between the two groups with regard to sleeping patterns or return to normal diet. The results of the main study are waited with interest. Ketoprofen has also been used during tonsillectomy. As shown with other NSAIDs, the analgesic effect of ketoprofen alone was not sufficient, but combined with an opioid or acetaminophen, ketoprofen provided adequate pain relief during recovery, with a low incidence of PONV and postoperative bleeding. [53,54] Pain after tonsillectomy is significant and, therefore, NSAIDs do not provide sufficient analgesia alone. Hence, NSAIDs should be combined with acetaminophen and/or opioids to provide satisfactory pain relief after surgery. [38,54,93,132] When used in appropriate doses, and when the first dose of NSAID is given after surgery, it seems that the incidence of postoperative bleeding is not increased. [49,50] However, caution should be used in treating children with any known pre-existing bleeding disorder. 5.5 Dental Surgery Most dental procedures are performed in the dentist s office using local anesthesia, but young and mentally retarded children may require general anesthesia. Most children should be administered an analgesic to provide calm recovery. Purday et al. [76] compared ketorolac with morphine in 120 children, aged between 2 and 10 years, undergoing dental surgery. In the ketorolac-treated group, two-thirds of the children needed rescue analgesia in the recovery room compared with one-third in the morphine-treated group, but the difference between the groups was not statistically significant. After discharge, 43% of children in the ketorolac-treated group needed analgesics at home compared with 23% of children in the morphine-treated group, but the difference between the groups was again not statistically significant. PONV occurred significantly more frequently in the morphine recipients than in the ketorolac recipients. Two studies demonstrate that ibuprofen performs better than placebo, and at least as well as acetaminophen, in children undergoing dental procedures. [93,132] 5.6 Strabismus Pain following ophthalmic surgery in children has been given less attention than PONV following the same procedures. Although eye surgery may be considered as minor surgery, ocular pain is common and may be extremely unpleasant. [133] Therefore, most children should be provided with analgesic treatment for 2 3 days after surgery. Opioids are often withheld considering the high incidence of PONV associated with squint surgery. Among children treated with opioids, when prophylactic antiemetic treatment is not used, the incidence of PONV varies from 46 88%. [134] Topical NSAID eye drops have been used in the symptomatic treatment of pain after strabismus surgery (table III). Topical NSAIDs have the theoretical advantage of avoiding the risk of systemic adverse effects. Snir et al. [86] compared diclofenac eye drops with dexamethasone eye drops in 40 children aged between 2 and 17 years. The children in the diclofenac-treated group had significantly lower pain scores during the 4-weeks follow-up than those in the dexamethasone-treated group. In another study, diclofenac eye drops provided similar analgesia to oxybuprocaine eye drops. [135] In contrast to promising results with diclofenac, ketorolac eye drops did not decrease pain scores, requirement for supplementary analgesia, or PONV, compared with placebo. [136] Both intramuscular [57] and intravenous [98] ketoprofen perform significantly better than placebo and meperidine (pethidine) in postoperative pain treatment after strabismus surgery. Moreover, in these two randomized, double-blind trials, the incidence of PONV was significantly less in the patients treated with ketoprofen than in the control groups. Three randomized, double-blind trials [ ] have compared ketorolac with different opioids given at induction of anesthesia. No differences between the groups were found in pain scores or the need for rescue analgesia, but the incidence and severity of PONV was significantly less in children receiving ketorolac (17 23%) compared with the opioid-treated groups (71 73%). Flurbiprofen, at a dose of 1 mg/kg but not at a dose of 0.5 mg/kg, provides a significant analgesic action after strabismus surgery, but does not decrease the incidence of PONV. [77] Nishina et al. [72] compared the efficacy of preoperative rectal diclofenac 2 mg/kg, intravenous flurbiprofen 1 mg/kg, and oral clonidine 4 µg/kg, given alone, as well as the combination of

11 NSAIDs for Postoperative Pain 113 ((Insert Table III here)) diclofenac plus clonidine, and flurbiprofen plus clonidine in controlling postoperative pain in 125 children. No significant differences were observed between the two NSAIDs in the pain scores or PONV. However, the recovery of children who were provided both clonidine and one of the two NSAIDs was more peaceful than those receiving NSAID monotherapy. In a recent non-blind, randomized study, [68] rectally administered diclofenac and intravenous morphine provided equal analgesic effect. However, it should be noted that acetaminophen was given to every patient. The incidence of PONV was only 12% in children treated with diclofenac plus acetaminophen compared with 72% in patients treated with morphine plus acetaminophen. In conclusion, several studies report a PONV incidence of <20% following pediatric strabismus surgery when proactive pain treatment with NSAIDs has been used. Therefore, NSAIDs may be particularly useful analgesics during strabismus surgery. 5.7 Herniotomy Peripheral nerve blocks and wound infiltration with local anesthetic are highly effective in the treatment of pain after inguinal surgery in children. However, regional anesthesia should be supplemented with an analgesic to ensure continuing pain relief, because postoperative pain commonly lasts 2 or 3 days after surgery and 80% of children need analgesic treatment after discharge (usually 1 9 doses). [61] Two studies [87,88] show that diclofenac performs at least as effectively as caudal epidural with bupivacaine in pain relief after herniotomy. However, the combination of appropriate regional analgesia with diclofenac seems to perform even better than each method alone. [89] Bean-Lijewski and Hunt [109] demonstrated a similar analgesic action with intramuscular ketorolac and intramuscular meperidine in children aged between 1 and 11 years undergoing herniorrhaphy or hydrocelectomy. The analgesic action of both drugs was significantly better than that of placebo, but no differences were noted between the three groups with regard to adverse effects. Two studies support the expectations that coadministration of ketorolac with regional analgesia may be beneficial in children. Graham and Wandless [110] found no statistically significant additional analgesic effect while combining a single intramuscular dose of ketorolac as an adjuvant to wound infiltration with bupivacaine in children undergoing umbilical hernia surgery. However, the need for rescue analgesia was less in the ketorolac-treated group compared with the control group. Splinter et al. [111] compared intravenous ketorolac with epidural bupivacaine-epinephrine (adrenaline) in children undergoing herniotomy. All children received a field block with bupivacaine.