Comparative effect of paracetamol, NSAIDs or their combination in postoperative pain management: a qualitative review

M. Hyllested1, S. Jones2, J. L. Pedersen1 and H. Kehlet*,1

1Department of Surgical Gastroenterology, Hvidovre University Hospital, Kettegaard Allé 30, DK-2650 Hvidovre, Denmark. 2Department of Anaesthesia, Waikato Hospital, Hamilton, New Zealand*Corresponding author

Accepted for publication: July 3, 2001


    Abstract
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Background. Quantitative reviews of postoperative pain management have demonstrated that the number of patients needed to treat for one patient to achieve at least 50% pain relief (NNT) is 2.7 for ibuprofen (400 mg) and 4.6 for paracetamol (1000 mg), both compared with placebo. However, direct comparisons between paracetamol and non-steroidal anti-inflammatory drugs (NSAIDs) have not been extensively reviewed. The aims of this review are (i) to compare the analgesic and adverse effects of paracetamol with those of other NSAIDs in postoperative pain, (ii) to compare the effects of combined paracetamol and NSAID with those of either drug alone, and (iii) to discuss whether the adverse effects of NSAIDs in short-term use are justified by their analgesic effects, compared with paracetamol.

Methods. Medline (1966 to January 2001) and the Cochrane Library (January 2001) were used to perform a systematic, qualitative review of postoperative pain studies comparing paracetamol (minimum 1000 mg) with NSAID in a double-blind, randomized manner. A quantitative review was not performed as too many studies of high scientific standard (27 out of 41 valid studies, including all major surgery studies) would have been excluded.

Results. NSAIDs were clearly more effective in dental surgery, whereas the efficacy of NSAIDs and paracetamol seemed without substantial differences in major and orthopaedic surgery, although firm conclusions could not be made because the number of studies was limited. The addition of an NSAID to paracetamol may confer additional analgesic efficacy compared with paracetamol alone, and the limited data available also suggest that paracetamol may enhance analgesia when added to an NSAID, compared with NSAIDs alone.

Conclusion. Paracetamol is a viable alternative to the NSAIDs, especially because of the low incidence of adverse effects, and should be the preferred choice in high-risk patients. It may be appropriate to combine paracetamol with NSAIDs, but future studies are required, especially after major surgery, with specific focus on a potential increase in side-effects from their combined use.

Br J Anaesth 2002; 88: 199–214

Keywords: analgesics non-opioid, paracetamol; analgesics anti-inflammatory, non-steroidal; pain, postoperative; analgesics


    Introduction
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
The aims of this review are (i) to compare the analgesic and adverse effects of paracetamol with those of other non-steroidal anti-inflammatory drugs (NSAIDs) in postoperative pain, (ii) to compare the effects of paracetamol–NSAID combination with those of either drug alone, and (iii) to discuss whether the adverse effects of NSAIDs in short-term use are justified by their analgesic effects compared with paracetamol. In recent systematic quantitative reviews of postoperative pain management based on placebo-controlled trials, ibuprofen 400 mg was shown to have a number needed to treat (NNT) of 2.7 compared with placebo,1 whereas paracetamol had an NNT of 4.6.2 NNT is the number of patients needed to treat for one patient to achieve at least 50% pain relief. However, the analgesic effect of NSAIDs vs paracetamol assessed in direct comparisons has not been reviewed extensively before.35


    Methods
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
A systematic review of the literature using Medline (1966 to January 2001) and the Cochrane Library (January 2001) was performed. The search profile included a comprehensive list of pain terms combined with ‘paracetamol’, ‘acetaminophen’, ‘propacetamol’, ‘non-steroidal anti-inflammatory drugs (NSAID)’ or individual drug names. Additional papers not indexed in the databases mentioned were retrieved by reviewing the reference lists from the published material.

Inclusion criteria were postoperative pain, double-blind design, randomized allocation, studies on man, English language and full journal publication. The statistical method had to be described in the study. Each report meeting the inclusion criteria was read by two of the authors and scored for inclusion and methodological quality using a three-item scale of 1–5.6 Two of the authors agreed on the scores. Reports described as randomized were given 1 point and an additional point if the method of randomization was described and it was appropriate (table of random numbers, computer-generated coin-tossing). Conversely, 1 point was deducted if the method of randomization was inappropriate (alternative allocation, allocation according to date of birth). One point was given when the study was described as double-blind and an additional point if the method of double-blinding was described and was appropriate (identical placebo, dummy). Again, 1 point was deducted if the blinding was inappropriate. Finally, 1 point was given to studies with a description of withdrawals and dropouts. Studies without randomization and blinding were excluded from the review, so the minimum score of an included trial was 2 and the maximum score 5. The studies did not have to be placebo-controlled as the analgesic effect of paracetamol and NSAIDs compared with placebo has been established.1 2 Clinical trials comparing paracetamol with NSAIDs [including acetylsalicylic acid (ASA)] were sought, as were studies evaluating paracetamol added to an NSAID against paracetamol or NSAID alone.

The dose of paracetamol had to be a minimum of 1000 mg when given as a single agent, because doses below 1000 mg may be insufficient.7 However, studies employing lower doses of paracetamol were included when given in combination with another NSAID or when administered to children. A wide range of NSAID doses was included. The medication could be administered at different times, including pre- and postoperatively, and by different routes such as i.v., oral and rectal.

Analgesic efficacy was evaluated by significant differences in standard pain measures and/or consumption of opioids/rescue analgesia.


    Results
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
A detailed description of all the studies is presented in Tables 13. The patient numbers in the tables excluded those receiving placebo, as it was the numbers receiving paracetamol and/or NSAIDs that we sought to evaluate. The studies were divided into the following comparisons: paracetamol vs NSAIDs (Table 1), paracetamol with NSAIDs vs paracetamol (Table 2) and paracetamol with NSAIDs vs NSAIDs (Table 3). The tables were subdivided into major and minor surgery. Some of the studies belonged to several categories and are therefore mentioned more than once.


View this table:
[in this window]
[in a new window]
 
Table 1 Paracetamol vs NSAIDs in postoperative pain. n refers to the number of patients involved in the specific comparisons, not total number of patients in the study. Analgesic outcome results for paracetamol vs NSAID: ‘{uparrow} = greater effect’ means that paracetamol was more effective than NSAID; ‘-> = same effect’ means that paracetamol and NSAID had the same effect; ‘{downarrow} = less effect’ means that paracetamol was less effective than NSAID. Analgesic outcome results were quantified when possible (e.g. VAS scores, rescue medication, PCA). Ordinal scale measures cannot be used for quantitative comparisons. OPS = objective and behavioural pain score; P = paracetamol; S = suppository
 

View this table:
[in this window]
[in a new window]
 
Table 3 Paracetamol combined with NSAIDs vs NSAIDs in postoperative pain. n refers to the number of patients involved in the specific comparisons, not total number of patients in the study. Analgesic outcome results for paracetamol and NSAID vs NSAID: ‘{uparrow} = greater effect’ means that the combination was better than NSAID alone; ‘-> = same effect’ means that the combination had the same effect as NSAID alone; ‘{downarrow} = less effect’ means that the combination was less effective than NSAID alone. Analgesic outcome results were quantified when possible (e.g. VAS scores, rescue medication, PCA). Ordinal scale measures cannot be used for quantitative comparisons. P = paracetamol; Supp = suppository
 

View this table:
[in this window]
[in a new window]
 
Table 2 Paracetamol combined with NSAIDs vs paracetamol in postoperative pain. n refers to the number of patients involved in the specific comparisons, not the total number of patients in the study. Analgesic outcome results for paracetamol and NSAID vs paracetamol: ‘{uparrow} = greater effect’ means that the combination was better than paracetamol alone; ‘-> = same effect’ means that the combination had the same effect as paracetamol alone; ‘{downarrow} = less effect’ means that the combination was less effective than paracetamol alone. Analgesic outcome results were quantified when possible (e.g. VAS scores, rescue medication, PCA). Ordinal scale measures cannot be used for quantitative comparisons. P = paracetamol; Supp = suppository
 
We found a total of 47 double-blind and randomized studies, of which six had to be excluded because of inadequate randomization (consecutive allocation) or inadequate statistical methods.813 Three further studies were excluded from evaluation because they failed to demonstrate statistically significant differences in pain scores or opioid consumption between groups receiving drugs of known analgesic efficacy and placebo controls, thus suggesting that the studies lacked sensitivity.1416 However, these studies are presented in the tables as they exhibit no apparent methodological problems and separation from placebo does not indicate ability to demonstrate a difference between active drugs. Several studies without placebo controls were included in the review, but these studies are considered to provide weaker evidence when no significant differences between active drugs were demonstrated.

Methodological quality scores ranged from 2 to 5 for all studies. The median value of quality scores for the positive studies (the studies which showed a difference in analgesic effect) and the negative studies were both 4. No statistical difference was found between the two groups using Mann–Whitney test (P=1.0).

Paracetamol vs NSAIDs
There were a total of 36 studies including 3362 patients undergoing a wide variety of surgical procedures (Table 1).

Major surgery
There were four valid studies in major abdominal and gynaecological surgery1720 and one involving laparoscopic cholecystectomy,21 including a total of 398 patients. In the most robust study,20 rectal diclofenac 50 mg was superior to rectal paracetamol regarding pain scores, but resulted in an equivalent morphine-sparing effect (36 and 40% respectively). There were no significant differences between paracetamol and NSAIDs in pain scores or postoperative morphine requirement in the other four studies. However, there were problems in these studies. Montgomery and colleagues17 studied a single rectal dose of diclofenac or paracetamol administered preoperatively and assessed its efficacy over 24 h. However, there were significant differences in age and body mass index between the groups, which could have affected the opioid requirements. Witjes and colleagues19 used a relatively insensitive four-point pain scale and consumption of buprenorphine tablets as efficacy measures and found no differences in pain scores between active medication and placebo, but a reduction (P=0.048) in opioid consumption on the day of surgery (consumption of buprenorphine tablets: placebo group 2.3, paracetamol group 1.5, naproxen group 1.8). There were no differences between NSAIDs, paracetamol and placebo on the subsequent 2 days, suggesting low study sensitivity. Out of the three best studies,18 20 21 only one had a placebo control20 and therefore proven sensitivity. In this study,20 diclofenac was superior to paracetamol regarding pain scores. The two other studies18 21 showed no significant difference in pain scores and none of the three studies showed differences in opioid requirement. In three studies,17 19 20 paracetamol was administered rectally, which may give lower bioavailability.3 High bioavailability of paracetamol was present in two out of the three best studies, as paracetamol was administered orally in the study of Owen and colleagues21 and i.v. in the study of Varrassi and colleagues.18 In these studies, there were no significant differences in pain scores or opioid consumption between paracetamol and NSAIDs. In summary, the limited number of studies with an optimal design precludes firm conclusions about a potential difference in analgesic effect between paracetamol and NSAIDs in major abdominal surgery. So far, the studies failed to show a substantial difference in analgesic efficacy between paracetamol and NSAIDs.

Orthopaedic surgery
Three trials including 270 orthopaedic patients were analysed.2224 None showed any differences in pain scores at rest. However, in one study evaluating pain on movement after disc surgery, ketoprofen was superior.24 In two robust studies with proven sensitivity, both employing 1000 mg oral doses of paracetamol, McQuay and colleagues found lower opioid requirements after bromfenac 25 mg but not 10 mg compared with paracetamol,23 but no difference when paracetamol was compared with ketorolac 10–20 mg.22 In summary, three studies have shown that the efficacy of paracetamol was not substantially different from that of NSAIDs, but again the limited number of studies precludes firm conclusions about the potential difference between paracetamol and NSAIDs. Paracetamol was administered orally or i.v. in all studies avoiding the more unpredictable bioavailability associated with the rectal route.

Gynaecological surgery
There were three trials involving a total of 178 patients after episiotomy25 26 (103 patients) or tubal occlusion27 (75 patients). In two placebo-controlled studies, ibuprofen (400 mg)25 and meclofenamate (100 and 200 mg)27 improved pain scores compared with paracetamol, but no differences in rescue medication were demonstrated. In the third study, which included only 30 patients, paracetamol was equivalent to naproxen 500 mg but the study sensitivity was not proven.26 In summary, NSAID was superior to paracetamol in two assay-sensitive trials involving two different surgical procedures.

Ear, nose and throat surgery
There were six valid studies that involved a total of 408 children undergoing ear, nose and throat surgery (myringotomy, adenoidectomy, tonsillectomy).2833 One study showed ketorolac (1 mg kg–1) to be superior to paracetamol (10 mg kg–1)32 and paracetamol equal to placebo, possibly reflecting the low dose. Four other studies showed that diclofenac29 30 33 and ketorolac31 were equivalent to paracetamol concerning objective pain scores and visual analogue scale (VAS) scores. In the study of Bean-Lijewski and Stinson,28 there was no clear conclusion. In three out of the six studies, no comparison of opioid requirements could be made.28 30 32 Opioid requirements were lowered by diclofenac in one study29 but equivalent to paracetamol in two other studies involving diclofenac and ketorolac.31 33 In a study of tonsillectomy, rectal paracetamol (35 mg kg–1) was equivalent to ketorolac (1 mg kg–1) i.v. despite all serum concentrations of paracetamol being below the antipyretic level.31 However, even when high doses of oral paracetamol (90 mg kg–1 per 24 h) were given to children after tonsillectomy, this did not improve analgesia compared with diclofenac (2–3 mg kg–1 per 24 h).30 Only one of these six studies included a placebo control.32 There are problems in interpreting these studies because pain rating in children is difficult. Five out of six studies included no placebo control2831 33 and could not differentiate between paracetamol and NSAID. In the study with a placebo control, ketorolac was superior to a relatively low dose of paracetamol (10 mg kg–1).32

Dental surgery
Of 16 dental studies, eight showed that NSAIDs were superior to paracetamol with respect to pain scores (1329 patients),3441 five showed that they were equivalent (370 patients)4246 and two that paracetamol 1000 mg was superior to aspirin 650 mg47 and diclofenac 100 mg.48 One study was not evaluated as the statistical comparison of paracetamol with NSAIDs was not performed.49 Of the eight studies in which NSAIDs were superior regarding pain scores, three also showed NSAIDs to be superior regarding remedication (993 patients).36 38 39 Of the six studies showing no differences in pain scores, study sensitivity was unproven in three43 44 46 but the other three studies were robust.42 45 48 In one study, assay sensitivity was inferred because paracetamol plus codeine was superior to paracetamol.48 In this study, which involved 68 patients, paracetamol 1000 mg and diclofenac 100 mg were equivalent regarding total pain relief and summed pain intensity difference over 8 h but paracetamol was superior to diclofenac in the first 3 h postoperatively (P=0.001). This could be due to slow onset of action of the enteric-coated diclofenac preparation.48 Cooper and colleagues42 showed paracetamol 1000 mg to be equivalent to ketoprofen 100 mg and superior to ketoprofen 25 mg. Seymour and colleagues45 showed equivalence between paracetamol 1000 mg and ketoprofen 25 mg but did not examine a higher dose. In these three studies,42 45 48 there were also no differences in opioid requirements. In all dental studies the medication was given orally, thus making bioavailability comparable. In summary, NSAIDs seem to be superior to paracetamol in dental surgery, regarding both pain scores and remedication. Most of the studies in dental surgery were robust, with relatively sensitive pain measurement scales, adult patients and medication administered orally.

Summary
Out of 33 valid studies, three (all dental studies) showed that the NSAID was superior to paracetamol with respect to both pain scores and opioid requirement or remedication,36 38 39 two studies showed that NSAIDs reduced opioid requirement or remedication only compared with paracetamol,23 29 and 10 studies showed that analgesia was improved by NSAIDs compared with paracetamol regarding pain scores, but either did not report opioid requirement or remedication32 35 37 41 or found no differences.20 24 25 27 34 40 Sixteen studies showed no differences between paracetamol and NSAIDs in pain scores,1719 2123 26 2931 33 4246 and 10 of these studies also showed no differences in opioid requirement or remedication.1719 21 22 31 33 42 45 46 Five of these 16 studies19 22 23 42 45 showed significant differences between active drugs and placebo, strengthening their conclusion of no difference between NSAID and paracetamol. Two studies found paracetamol to be superior to NSAID regarding pain scores, but not remedication requirement.47 48 One study had an unclear conclusion28 and one study made no statistical comparison between paracetamol and NSAIDs.49

The efficacies of paracetamol and NSAIDs may depend on the type of surgery. Of the three best studies in major abdominal/gynaecological surgery (including laparoscopic cholecystectomy), two found no significant differences between paracetamol and NSAIDs18 21 and one demonstrated that NSAIDs were superior20 as regards pain scores. In all three studies, no significant difference was found in opioid requirement. However, there are several methodological problems in these studies and thus no clear conclusion can be made regarding the efficacy of NSAIDs and paracetamol in major surgery. In orthopaedic surgery, three robust studies showed that the efficacy of paracetamol was comparable to that of NSAIDs,2224 but more data are needed to allow final conclusions. In gynaecological minor surgery (episiotomy and laparoscopic tubal ligation), no clear conclusion could be made, but NSAIDs seemed to be more efficacious in two assay-sensitive studies. In ear, nose and throat surgery, no clear conclusion could be made but paracetamol and NSAIDs seemed equivalent. In dental surgery, NSAIDs seemed to be superior to paracetamol regarding pain scores and remedication requirements.

Thus, overall, NSAIDs seem to be superior to paracetamol in postoperative pain management, but the magnitude of the difference may depend on the type of surgery performed. In major surgery, the efficacies of NSAIDs and paracetamol seem to be comparable, whereas in minor surgery NSAIDs seem to be superior.

The combination of paracetamol and NSAID vs paracetamol alone
There was a total of eight studies, out of which seven could be included.17 24 48 5053 These involved 613 patients (Table 2). The last study failed to separate active drugs from placebo.14 Each study involved a different surgical procedure, making comparisons difficult. In four of the studies,24 48 50 52 the combinations of paracetamol with ASA, of paracetamol with ketoprofen 50 and 100 mg and of paracetamol with diclofenac were associated with lower pain scores than paracetamol alone. In the study of paracetamol and ketoprofen, the combination reduced pain scores both at rest and on movement after disc surgery compared with paracetamol alone.24 In a study involving spinal fusion surgery,52 the combination of propacetamol and ketoprofen 100 mg improved pain scores assessed by VAS pain intensity differences. Pain relief scores, on the other hand, were not significantly different between the two groups in this study. In two studies involving major gynaecological surgery,17 53 there were no differences in pain scores and in one study51 the pain scores were not measured.

In their assessment of opioid consumption, five out of the seven studies17 24 48 51 52 reported significant reductions, ranging from 33–46%, when both drugs were used compared with paracetamol alone. However, in one of the studies these findings may have been exaggerated by the low dose of paracetamol and demographic differences between the study groups, as discussed above.17 In the study involving spinal fusion surgery,52 the combination of propacetamol and ketoprofen reduced consumption of morphine under patient-controlled analgesia (PCA). In only one of the seven studies was there no advantage in adding an NSAID to paracetamol.53 This study compared high-dose rectal paracetamol (40 mg kg–1) with diclofenac 100 mg added to paracetamol (20 mg kg–1) and with paracetamol (20 mg kg–1) alone. The lack of difference between a low dose of paracetamol (20 mg kg–1) and its combination with a full dose of an NSAID may suggest low study sensitivity.

In summary, the addition of an NSAID to paracetamol seems to provide additional analgesic efficacy. However, whether this additional analgesic efficacy is a result of a true additive effect or a reflection of NSAIDs being more effective than paracetamol is not clear.

Paracetamol combined with NSAID vs NSAID alone
A total of five studies were found (Table 3), but only four of them, involving 190 patients, were included in our evaluation.17 24 48 54 One study was excluded because of failure to separate active drugs from placebo.14 In the most robust trial, the combination of propacetamol with ketoprofen 50 mg reduced pain scores at rest and on movement compared with ketoprofen alone after disc surgery, but there was no associated reduction in opioid requirement.24 Oral diclofenac 100 mg combined with paracetamol 1000 mg reduced pain intensity scores, improved pain relief scores and reduced the need for rescue analgesia compared with diclofenac alone after dental surgery,48 though this finding in part reflects the slow onset of an enteric-coated preparation. A dental surgery study54 found no differences between the combination of diclofenac with paracetamol and diclofenac alone, but the doses of diclofenac and paracetamol were only 50 and 500 mg respectively. In the remaining study, which involved elective gynaecological surgery,17 there were no significant differences between diclofenac alone and its combination with paracetamol in either pain scores or opioid requirement. However, this study17 has weaknesses because of differences in age and body mass index, as discussed above.

In summary, the available data are sparse but two trials suggest that standard doses of paracetamol do enhance analgesic efficacy when added to NSAIDs compared with NSAIDs alone.

Adverse effects of paracetamol vs NSAID
Relatively few studies have compared the adverse effects of NSAIDs and paracetamol, especially in the postoperative period. An exhaustive review of adverse effects is beyond the scope of this article but some important data regarding major adverse effects are presented together with a number of less well-known facts.

Gastrointestinal
Ultrastructural damage to the gastric surface epithelium occurs within minutes after ingestion of NSAIDs and gross endoscopically detectable haemorrhages and erosions in the gastroduodenal epithelium occur within hours.55 A review of short-term NSAID use concluded that there was no evidence of an increased risk of severe gastrointestinal complications during perioperative (<1 week) NSAID treatment.56 However, patients with active or previous gastroduodenal ulcer were excluded from most of the studies reviewed, and the risk of severe complications from short-term use of NSAIDs cannot be excluded in these patients.56 A study by Strom and colleagues,57 including 10 272 patients, showed that ketorolac was associated with a small increased risk of gastrointestinal bleeding (odds ratio=1.17) when analgesic therapy lasted for 5 or fewer days. However, the risk was significantly greater and clinically important when ketorolac was used in higher doses, in older patients and for more than 5 days.57 A multicentre study of 875 cases of upper gastrointestinal bleeding, verified by endoscopy, suggested that any use of aspirin for more than a 7-day period increased the risk of bleeding by about seven times, and that diclofenac, indometacin, naproxen and piroxicam were associated with a risk similar to that of aspirin.58 Paracetamol, propyphenazone and metamizole were not associated with this increased risk.

Allergic
NSAIDs may exacerbate asthma, especially in patients with aspirin-induced asthma.59 Settipane and colleagues60 determined the prevalence of cross-reactivity to high-dose paracetamol in 50 aspirin-sensitive asthmatic patients and in 20 non-aspirin-sensitive asthmatic control subjects. The study showed that non-aspirin sensitive asthmatic patients did not react to paracetamol, whereas in aspirin-sensitive patients 16 and 20% developed bronchospasm with paracetamol 1000 and 1500 mg respectively. The reactions were generally mild and easily reversed.

Hepatic
Overdose of paracetamol can occasionally lead to irreversible liver injury that can be lethal.61 The single adult dose that must be ingested to produce severe liver damage is about 150–250 mg kg–1, corresponding to a plasma concentration equal to or greater than 200 mg litre–1.62 Hepatotoxicity has been reported in chronic alcoholics after ingestion of therapeutic doses of paracetamol.63 However, paracetamol did not induce adverse effects in the liver in 20 patients with chronic liver disease (six with alcoholic liver disease) who were studied over 2 weeks in a double-blind cross-over design in which the patients were given paracetamol 4000 mg day–1 or placebo.64

Very rare hepatic injury has been observed for nearly all NSAIDs currently on the market, but diclofenac, sulindac and aspirin may be more commonly associated with liver disease.65

Renal
Prostaglandins have little influence on renal blood flow (RBF) or glomerular filtration rate (GFR) in normal healthy individuals66 but oppose the renal vasoconstriction induced by catecholamines, vasopressin and angiotensin in states such as hypovolaemia, congestive heart failure and cirrhosis with ascites.67 These conditions also prevail in many postoperative patients, who may have major shifts in fluid compartments as well as activation of the neurohumoral stress response. A recent meta-analysis of the influence of NSAIDs on the postoperative renal function of 183 patients with normal preoperative renal function found significantly reduced sodium and potassium excretion and 21–28% reduction in creatinine clearance on day 1 compared with controls. No significant differences were present on day 2 other than a mean rise in serum creatinine of 15 µmol litre–1.68

A retrospective cohort study69 found no evidence of an increased incidence of renal failure among 10 000 patients receiving postoperative ketorolac even in the presence of established risk factors, unless therapy exceeded 5 days, when the risk doubled. These conflicting sources of information are difficult to reconcile, but suggest that the readily demonstrable biochemical and haemodynamic effects do not often progress to an adverse outcome. Paracetamol exerts weaker inhibition of peripheral prostaglandin synthesis than NSAIDs.70 71 It does produce effects on sodium and water excretion comparable to those of NSAIDs,71 but not on RBF and GFR,71 even in the stressed kidney.72

Haematological
Most studies comparing the effects of NSAIDs and paracetamol on haemostasis have been performed in tonsillectomy patients. In two studies that involved a total of 1544 children treated with ASA 300–1000 mg or paracetamol 240–1000 mg, post-tonsillectomy haemorrhage was seen in 3.1–3.8% in the ASA group and 0.3–0.5% in the paracetamol group.73 74 Two studies compared blood loss after preoperative administration of non-ASA NSAIDs (rectal diclofenac 0.65–1.0 mg kg–1 or i.v. ketorolac 1 mg kg–1) and paracetamol. Both studies found significantly greater blood loss in patients receiving NSAIDs and significantly longer duration of surgery33 or a greater number of patients requiring additional measures to obtain haemostasis compared with paracetamol.31 Other prospective75 76 and retrospective77 78 studies have found increased postoperative bleeding in patients receiving perioperative ketorolac for tonsillectomy.

Miscellaneous
NSAIDs have significant inhibitory effects on heterotopic bone formation,79 whereas the effects on fracture union are debatable.80 However, similar studies on bone healing are not available for paracetamol.

Aspirin and ibuprofen have been shown to disrupt sleep compared with paracetamol and placebo. Thirty-seven male and female subjects had their sleep pattern recorded one night after ingestion of aspirin 650 mg, paracetamol 650 mg or ibuprofen 400 mg.81 Aspirin and ibuprofen disrupted sleep by increasing the number of awakenings and the percentage of time spent in stage wake and by decreasing sleep efficiency. Paracetamol did not differ significantly from placebo on any measure of the recorded sleep pattern.

Correspondingly, the normal decrease in nocturnal body temperature was attenuated and melatonin synthesis suppressed after NSAID compared with placebo administration in 75 subjects.82

Diclofenac has been shown to alter the pharmacokinetics of active morphine metabolites in patients with postoperative pain.83 Even though morphine consumption decreased by 20% after diclofenac was administered, the concentration of the active metabolite, morphine-6-glucuronide, was unchanged and a significant reduction in respiratory rate occurred.83


    Discussion
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Paracetamol was found to have analgesic efficacy comparable to that of NSAIDs in many of the studies reviewed, but overall, NSAIDs seem to be superior for postoperative pain management, although there seem to be differences in the efficacies of paracetamol and NSAIDs depending on the type of surgery performed. In major and orthopaedic surgery, the efficacies of NSAIDs and paracetamol seem to be comparable and in dental surgery NSAIDs seem superior.

Paracetamol and NSAIDs (ibuprofen and diclofenac) have been assessed compared with placebo in recent Cochrane systematic reviews.1 2 Paracetamol 1000 mg had an NNT of 4.6 compared with placebo, ibuprofen 400 mg had an NNT of 2.7 and diclofenac 50 mg an NNT of 2.3.1 2 In these Cochrane reviews, the NNT differences between paracetamol and NSAIDs were calculated from placebo-controlled studies in which dental studies constituted the majority. However, we cannot be certain whether these findings reflect inherent differences in efficacy between the drugs or differences in the sensitivity of the surgical models to NSAIDs and paracetamol. The NNT values may be misleading in the setting of moderate to major surgery, but the limited number of comparative studies in major surgery precludes final conclusions.

The opioid-sparing effect of NSAIDs has often been used as an analgesic efficacy parameter. However, recent studies have suggested that NSAIDs may reduce morphine requirements by reducing the excretion of the active metabolite, morphine-6-glucuronide. Morphine sparing cannot, therefore, be assumed to result in parallel reductions in opioid-related adverse effects. Fentanyl sparing may be a more appropriate surrogate end-point for future NSAID studies as this drug has minimal renal excretion and inactive metabolites.

The addition of NSAIDs to paracetamol may confer additional analgesic efficacy compared with paracetamol alone. Given the conclusion of the direct comparative studies—that NSAIDs may be more effective than paracetamol—the key question is whether the addition of paracetamol to an NSAID will be worthwhile in patients able to take either medication. Even though few robust data are available, standard doses of paracetamol may enhance analgesic efficacy when added to NSAIDs, compared with NSAIDs alone (two trials). Further evidence of this is seen in non-surgical studies of patients with rheumatoid arthritis, in whom indometacin (150 mg day–1) alone and the combination of indometacin (50 mg day–1) with paracetamol (4 g day–1) had the same analgesic effect, but the combination had fewer and milder side-effects.84 In two other studies, treatment with a combination of naproxen with paracetamol had a greater analgesic effect than treatment with higher naproxen doses alone.85 86 A review87 concerning paracetamol in rheumatoid arthritis suggests that there is increasing evidence that combined paracetamol and NSAID treatment is more effective than treatment with NSAIDs alone. The findings that the combination appears to be more effective than either drug alone may support the suggestion that NSAIDs are not greatly superior to paracetamol.

A formal quantitative review (meta-analysis) was not performed as too many studies of high scientific standard would have had to be discarded if we had used the method introduced by McQuay and Moore to convert different pain scales to a common denominator and thereby make them comparable.88 The key problem for many quantitative reviews is that a large number of papers must be discarded if they do not use standard scales of pain assessment, use analgesic drug consumption (e.g. PCA), employ pre-emptive techniques or involve local anaesthetic blocks. The next problem may be that the remaining trials are not representative.89 Of the valid studies in this review, 27 out of 41 (including all major surgery studies) would have had to be discarded if a quantitative review were to have been performed. There are also problems concerning qualitative reviews, as the simple vote-counting method may mislead. It ignores the sample size of the constituent studies, the magnitude of the effect in the studies and the validity of their design even when randomized.90 Inadequate or unclear randomization can overestimate the treatment effect by 30–41% and non-double-blind conditions can overestimate it by 17%.6 However, quality scores of the studies in this review had a median value of 4 on a scale of 1–5 and there were no differences in quality scores between studies showing a difference in analgesic effect and those that did not show such a difference.

A further perennial challenge for analgesic studies is the multidimensional and mutually opposed nature of the assessments: as pain improves patients are less likely to request analgesia, yet we seek statistically significant differences in one of these dimensions without attempting to anchor the other. Thus, in the study of Owen and colleagues,21 for example, the difference in pain scores between ibuprofen and paracetamol almost achieved statistical significance in favour of the NSAID (P=0.057) despite 50% lower opioid consumption in these patients, but as conventional levels of significance were not reached in either measure and as there was no placebo control this study was considered to represent weak evidence of equality. Had the opioid administration been fixed in both groups, one might surmise that there would have been a significant difference in pain scores, and thus strong evidence for the superiority of NSAIDs. Despite these problems, we consider that the comparison of paracetamol with NSAID in postoperative pain management is important, especially as the side-effects of these compounds are so different.

The very low apparent risk of paracetamol therapy suggests a highly favourable risk:benefit ratio, which might justify a role for paracetamol as a near-routine postoperative background analgesic. Where the additional analgesic effect of an NSAID is particularly sought, as after relatively minor or ambulatory surgery and when the perceived risks from NSAIDs are low, NSAIDs may be preferred as background analgesic.

There were bioavailability problems, especially in the major surgery and paediatric studies, as paracetamol was administered rectally, making the analgesic effect unpredictable compared with the oral or i.v. route. The pharmacokinetics of paracetamol has been reviewed recently, and the bioavailability of paracetamol given by the rectal route ranged from 24–98%.3 Serum and saliva concentrations after high-dose rectal and oral paracetamol were studied in postoperative adult patients,91 and it was concluded that administering paracetamol 2000 mg rectally resulted in serum and saliva concentrations during the first 4 h that never exceeded the minimum effective antipyretic serum concentration.

In conclusion, the existing direct comparative studies show that NSAIDs are more effective than paracetamol in some situations, e.g. dental surgery, but the differences are less obvious after other types of surgery. In many studies, paracetamol was given in insufficient doses or administered rectally, potentially underestimating the efficacy, whereas the reduction in morphine requirements may overestimate the inherent analgesic efficacy of the NSAIDs. Paracetamol is definitely a viable alternative to the NSAIDs, especially because of the lower incidence of adverse effects, and should be the preferred choice in high-risk patients. In the absence of firm data, paracetamol should also be considered instead of NSAIDs for pain management after major or orthopaedic surgery, as few differences in efficacies were found in existing data. After tonsillectomy, paracetamol is also recommended because of less bleeding. It may be appropriate to combine paracetamol with NSAIDs, but future studies are required, especially after major surgery. In such studies, there should also be a specific focus on a potential increase in side-effects from their combined use.


    References
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
1 Collins SL, Moore RA, McQuay HJ, Wiffen PJ, Edwards JE. Single dose oral ibuprofen and diclofenac for postoperative pain. Cochrane Systematic Rev 1998; 3 September: 1–10

2 Moore A, Collins S, Carroll D, McQuay H, Edwards J. Single dose paracetamol (acetaminophen), with and without codeine, for postoperative pain. Cochrane Systematic Rev 1998; 18 August: 1–9

3 Ward B, Alexander-Williams JM. Paracetamol revisited: a review of the pharmacokinetics and pharmacodynamics. Acute Pain 1999; 2: 139–49

4 Wiebalck CA, Van Aken H. Paracetamol and propacetamol for post-operative pain: contrasts to traditional NSAIDs. Baillière’s Clin Anaesthesiol 1995; 9: 469–82

5 Ahmad N, Grad HA, Haas DA, Aronson KJ, Jokovic A, Locker D. The efficacy of nonopioid analgesics for postoperative dental pain: a meta-analysis. Anesth Prog 1997; 44: 119–26[Medline]

6 McQuay H, Moore A. Judging the quality of trials. In: McQuay H, Moore A, eds. An Evidence-Based Resource for Pain Relief. Oxford: Oxford University Press, 1998; 10–3

7 Moore A, Collins S, Carroll D, McQuay H. Paracetamol with and without codeine in acute pain: a quantitative systematic review. Pain 1997; 70: 193–201[ISI][Medline]

8 Gomez-Jimenez J, Franco-Patino R, Chargoy-Vera J, Olivares-Sosa R. Clinical efficacy of mild analgesics in pain following gynaecological or dental surgery: report on multicentre studies. Br J Clin Pharmacol 1980; 10 Suppl 2: 355S–358S[Medline]

9 Breivik H, Spilsberg AM, Winnem B, Apalseth K. Relief of postoperative pain with zomepirac, paracetamol with codeine, or placebo. In: Vickers MD, Lunn JN, eds. Mortality in Anaesthesia. Berlin: Springer-Verlag, 1983; 196–207

10 Parkhouse J, Hallinon P. A comparison of aspirin and paracetamol. Br J Anaesth 1967; 39: 146–54[ISI][Medline]

11 Muckle DS. Open meniscectomy: enhanced recovery after synovial prostaglandin inhibition. J Bone Joint Surg Br 1984; 66: 193–5[ISI][Medline]

12 Davie IT, Gordon NH. Comparative assessment of fenoprofen and paracetamol given in combination for pain after surgery. Br J Anaesth 1978; 50: 931–5[Abstract]

13 Torabinejad M, Cymerman JJ, Frankson M, Lemon RR, Maggio JD, Schilder H. Effectiveness of various medications on postoperative pain following complete instrumentation. J Endodon 1994; 20: 345–54[ISI][Medline]

14 VanLancker P, Vandekerckhove B, Cooman F. The analgesic effect of preoperative administration of propacetamol, tenoxicam or a mixture of both in arthroscopic, outpatient knee surgery. Acta Anaesthesiol Belg 1999; 50: 65–9[Medline]

15 Dahl V, Ernø PE, Ræder JC. No analgesic effect of ibuprofen or paracetamol vs placebo for hysterectomies. Eur J Pain 1997; 1: 31–5

16 Derkay CS, Wadsworth JT, Darrow DH, Strasnick B, Thompson GK, O’Master J. Tube placement: a prospective, randomized double-blind study. Laryngoscope 1998; 108: 97–101[ISI][Medline]

17 Montgomery JE, Sutherland CJ, Kestin IG, Sneyd JR. Morphine consumption in patients receiving rectal paracetamol and diclofenac alone and in combination. Br J Anaesth 1996; 77: 445–7[Abstract/Free Full Text]

18 Varrassi G, Marinangeli F, Agro F et al. A double-blinded evaluation of propacetamol versus ketorolac in combination with patient-controlled analgesia morphine: analgesic efficacy and tolerability after gynecologic surgery. Anesth Analg 1999; 88: 611–6[Abstract/Free Full Text]

19 Witjes WP, Crul BJ, Vollaard EJ, Joosten HJ, von Egmond J. Application of sublingual buprenorphine in combination with naproxen or paracetamol for post-operative pain relief in cholecystectomy patients in a double-blind study. Acta Anaesthesiol Scand 1992; 36: 323–7[ISI][Medline]

20 Cobby TF, Crighton IM, Kyriakides K, Hobbs GJ. Rectal paracetamol has a significant morphine-sparing effect after hysterectomy. Br J Anaesth 1999; 83: 253–6[Abstract/Free Full Text]

21 Owen H, Plummer JL, Ilsley AH, Tordoff K, Toouli J. Pain control in the week following laparoscopic surgery: a comparison of sustained-release ibuprofen and paracetamol. Min Invasive Ther Allied Technol 1997; 6: 235–40

22 McQuay HJ, Poppleton P, Carroll D, Summerfield RJ, Bullingham RE, Moore RA. Ketorolac and acetaminophen for orthopedic postoperative pain. Clin Pharmacol Ther 1986; 39: 89–93[ISI][Medline]

23 McQuay HJ, Carroll D, Frankland T, Harvey M, Moore A. Bromfenac, acetaminophen, and placebo in orthopedic postoperative pain. Clin Pharmacol Ther 1990; 47: 760–6[ISI][Medline]

24 Fletcher D, Negre I, Barbin C, Francois A, Carreres C, Falgueirettes C, Barboteu A, Samii K. Postoperative analgesia with i.v. propacetamol and ketoprofen combination after disc surgery. Can J Anaesth 1997; 44: 479–85[Abstract]

25 Schachtel BP, Thoden WR, Baybutt RI. Ibuprofen and acetaminophen in the relief of postpartum episiotomy pain. J Clin Pharmacol 1989; 29: 550–3[Abstract/Free Full Text]

26 Skovlund E, Fyllingen G, Landre H, Nesheim BI. Comparison of postpartum pain treatments using a sequential trial design: II. Naproxen versus paracetamol. Eur J Clin Pharmacol 1991; 40: 539–42[ISI][Medline]

27 Huang KC, Wolfe WM, Tsueda K, Simpson PM, Caissie KF. Effects of meclofenamate and acetaminophen on abdominal pain following tubal occlusion. Am J Obstet Gynecol 1986; 155: 624–9[ISI][Medline]

28 Bean-Lijewski JD, Stinson JC. Acetaminophen or ketorolac for post myringotomy pain in children? A prospective, double-blinded comparison. Paediatr Anaesth 1997; 7: 131–7[ISI][Medline]

29 Baer GA, Rorarius MG, Kolehmainen S, Selin S. The effect of paracetamol or diclofenac administered before operation on postoperative pain and behaviour after adenoidectomy in small children. Anaesthesia 1992; 47: 1078–80[ISI][Medline]

30 Rømsing J, Østergaard D, Drozdziewicz D, Schultz P, Ravn G. Diclofenac or acetaminophen for analgesia in paediatric tonsillectomy outpatients. Acta Anaesthesiol Scand 2000; 44: 291–5[ISI][Medline]

31 Rusy LM, Houck CS, Sullivan LJ et al. A double-blind evaluation of ketorolac tromethamine versus acetaminophen in pediatric tonsillectomy: analgesia and bleeding. Anesth Analg 1995; 80: 226–9[Abstract]

32 Watcha MF, Ramirez Ruiz M, White PF, Jones MB, Lagueruela RG, Terkonda RP. Perioperative effects of oral ketorolac and acetaminophen in children undergoing bilateral myringotomy. Can J Anaesth 1992; 39: 649–54[Abstract]

33 Schmidt A, Björkman S, Åkeson J. Preoperative rectal diclofenac versus paracetamol for tonsillectomy: effects on pain and blood loss. Acta Anaesthesiol Scand 2001; 45: 48–52[ISI][Medline]

34 Cooper SA, Schachtel BP, Goldman E, Gelb S, Cohn P. Ibuprofen and acetaminophen in the relief of acute pain: a randomized, double-blind, placebo-controlled study. J Clin Pharmacol 1989; 29: 1026–30[Abstract/Free Full Text]

35 Dionne RA. Suppression of dental pain by the preoperative administration of flurbiprofen. Am J Med 1986; 80: 41–9

36 Kiersch TA, Halladay SC, Hormel PC. A single-dose, double-blind comparison of naproxen sodium, acetaminophen, and placebo in postoperative dental pain. Clin Ther 1994; 16: 394–404[ISI][Medline]

37 McGaw T, Raborn W, Grace M. Analgesics in pediatric dental surgery: relative efficacy of aluminum ibuprofen suspension and acetaminophen elixir. ASDC J Dent Child 1987; 54: 106–9[ISI][Medline]

38 Mehlisch DR, Sollecito WA, Helfrick JF et al. Multicenter clinical trial of ibuprofen and acetaminophen in the treatment of postoperative dental pain. J Am Dent Assoc 1990; 121: 257–63[ISI][Medline]

39 Mehlisch DR, Jasper RD, Brown P, Korn SH, McCarroll K, Murakami AA. Comparative study of ibuprofen lysine and acetaminophen in patients with postoperative dental pain. Clin Ther 1995; 17: 852–60[ISI][Medline]

40 Nystrom E, Gustafsson I, Quiding H. The pain intensity at analgesic intake, and the efficacy of diflunisal in single doses and effervescent acetaminophen in single and repeated doses. Pharmacotherapy 1988; 8: 201–9[ISI][Medline]

41 Olstad OA, Skjelbred P. The effects of indoprofen vs paracetamol on swelling, pain and other events after surgery. Int J Clin Pharmacol Ther Toxicol 1986; 24: 34–8[Medline]

42 Cooper SA, Reynolds DC, Reynolds B, Hersh EV. Analgesic efficacy and safety of (R)-ketoprofen in postoperative dental pain. J Clin Pharmacol 1998; 38: 11S–18S[Abstract/Free Full Text]

43 Irvine GH, Lutterloch MJ, Bowerman JE. Comparison of diflunisal and paracetamol in the management of pain following wisdom teeth removal. Br Dent J 1982; 152: 18–20[ISI][Medline]

44 Rodrigo C, Chau M, Rosenquist J. A comparison of paracetamol and diflunisal for pain control following 3rd molar surgery. Int J Oral Maxillofac Surg 1989; 18: 130–2[ISI][Medline]

45 Seymour RA, Kelly PJ, Hawkesford JE. The efficacy of ketoprofen and paracetamol (acetaminophen) in postoperative pain after third molar surgery. Br J Clin Pharmacol 1996; 41: 581–5[ISI][Medline]

46 Cheung LK, Rodrigo C. Tenoxicam for pain relief following third molar surgery. Anesth Pain Control Dent 1992; 4: 229–33[Medline]

47 Mehlisch DR, Frakes LA. A controlled comparative evaluation of acetaminophen and aspirin in the treatment of postoperative pain. Clin Ther 1984; 7: 89–97[ISI][Medline]

48 Breivik EK, Barkvoll P, Skovlund E. Combining diclofenac with acetaminophen or acetaminophen-codeine after oral surgery: a randomized, double-blind single-dose study. Clin Pharmacol Ther 1999; 66: 625–35[ISI][Medline]

49 Moore PA, Acs G, Hargreaves JA. Postextraction pain relief in children: a clinical trial of liquid analgesics. Int J Clin Pharmacol Ther Toxicol 1985; 23: 573–7[Medline]

50 Rubin A, Winter L Jr. A double-blind randomized study of an aspirin/caffeine combination versus acetaminophen/aspirin combination versus acetaminophen versus placebo in patients with moderate to severe post-partum pain. J Int Med Res 1984; 12: 338–45[ISI][Medline]

51 Mather SJ, Peutrell JM. Postoperative morphine requirements, nausea and vomiting following anaesthesia for tonsillectomy. Comparison of intravenous morphine and non-opioid analgesic techniques. Paediatr Anaesth 1995; 5: 185–8[ISI][Medline]

52 Aubrun F, Langeron O, Heitz D, Coriat P, Riou B. Randomised, placebo-controlled study of the postoperative analgesic effects of ketoprofen after spinal fusion surgery. Acta Anaesthesiol Scand 2000; 44: 934–9[ISI][Medline]

53 Beck DH, Schenk MR, Hagemann K, Doepfmer UR, Kox WJ. The pharmacokinetics and analgesic efficacy of larger dose rectal acetaminophen (40 mg/kg) in adults: a double-blinded, randomized study. Anesth Analg 2000; 90: 431–6[Abstract/Free Full Text]

54 Matthews RW, Scully CM, Levers BG. The efficacy of diclofenac sodium (Voltarol) with and without paracetamol in the control of post-surgical dental pain. Br Dent J 1984; 157: 357–9[ISI][Medline]

55 Wolfe MM, Lichtenstein DR, Singh G. Gastrointestinal toxicity of nonsteroidal antiinflammatory drugs. N Engl J Med 1999; 340: 1888–99[Free Full Text]

56 Kehlet H, Dahl JB. Are perioperative nonsteroidal anti-inflammatory drugs ulcerogenic in the short term? Drugs 1992; 44: 38–41[Medline]

57 Strom BL, Berlin JA, Kinman JL et al. Parenteral ketorolac and risk of gastrointestinal and operative site bleeding. JAMA 1996; 275: 376–82[Abstract]

58 Laporte JR, Carne X, Vidal X, Moreno V, Juan J. Upper gastrointestinal bleeding in relation to previous use of analgesics and non-steroidal anti-inflammatory drugs. Catalan Countries Study on Upper Gastrointestinal Bleeding. Lancet 1991; 337: 85–9[ISI][Medline]

59 Power, I. Aspirin-induced asthma. Br J Anaesth 1993; 71: 619–20[ISI][Medline]

60 Settipane RA, Schrank PJ, Simon RA, Mathison DA, Christiansen SC, Stevenson DD. Prevalence of cross-sensitivity with acetaminophen in aspirin-sensitive asthmatic subjects. J Allergy Clin Immunol 1995; 96: 480–5[ISI][Medline]

61 Prescott LF, Roscoe P, Wright N, Brown SS. Plasma-paracetamol half-life and hepatic necrosis in patients with paracetamol overdosage. Lancet 1971; March 13: 519–22

62 Rainsford KD, Roberts SC, Brown S. Ibuprofen and paracetamol: relative safety in non-prescription dosages. J Pharm Pharmacol 1997; 49: 345–76[ISI][Medline]

63 Seeff LB, Cuccherini BA, Zimmerman HJ, Adler E, Benjamin SB. Acetaminophen hepatotoxicity in alcoholics. A therapeutic misadventure. Ann Intern Med 1986; 104: 399–404[ISI][Medline]

64 Benson GD. Acetaminophen in chronic liver disease. Clin Pharmacol Ther 1983; 33: 95–101[ISI][Medline]

65 Bjorkman D. Nonsteroidal anti-inflammatory drug-associated toxicity of the liver, gastrointestinal tract, and esophagus. Am J Med 1998; 105: 17S–21S[Medline]

66 Haylor J. Prostaglandin synthesis and renal function in man. J Physiol (London) 1980; 298: 383–96[Abstract]

67 Murray MD, Brater DC. Renal toxicity of nonsteroidal anti-inflammatory drugs. Ann Rev Pharmacol Toxicol 1993; 32: 435–65

68 Lee A, Cooper MG, Craig JC, Knight JF, Keneally JP. The effects of nonsteroidal anti-inflammatory drugs (NSAIDs) on postoperative renal function: a meta-analysis. Anaesth Intensive Care 1999; 27: 574–80[ISI][Medline]

69 Feldman HI, Kinman JL, Berlin JA et al. Parenteral ketorolac: The risk of acute renal failure. Ann Intern Med 1997; 126: 193–9[Abstract/Free Full Text]

70 Berg KJ, Djoseland O, Gjellan A et al. Acute effects of paracetamol on prostaglandin synthesis and renal function in normal man and in patients with renal failure. Clin Nephrol 1990; 34: 255–62[ISI][Medline]

71 Prescott LF, Mattison P, Menzies DG, Manson LM. The comparative effects of paracetamol and indomethacin on renal function in healthy female volunteers. Br J Clin Pharmacol 1990; 29: 403–12[ISI][Medline]

72 Farquhar WB, Morgan AL, Zambraski EJ, Kenney WL. Effects of acetaminophen and ibuprofen on renal function in the stressed kidney. J Appl Physiol 1999; 86: 598–604[Abstract/Free Full Text]

73 Carrick DG. Salicylates and post-tonsillectomy haemorrhage. J Laryngol Otol 1984; 98: 803–5[ISI][Medline]

74 Stage J, Jensen JH, Bonding P. Post-tonsillectomy haemorrhage and analgesics. A comparative study of acetylsalicylic acid and paracetamol. Clin Otolaryngol 1988; 13: 201–4[ISI][Medline]

75 Gunter JB, Varughese AM, Harrington JF et al. Recovery and complications after tonsillectomy in children: a comparison of ketorolac and morphine. Anesth Analg 1995; 81: 1136–41[Abstract]

76 Splinter WM, Rhine EJ, Roberts DW, Reid CW, MacNeill HB. Preoperative ketorolac increases bleeding after tonsillectomy in children. Can J Anaesth 1996; 43: 560–3[Abstract]

77 Judkins JH, Dray TG, Hubbell RN. Intraoperative ketorolac and tonsillectomy bleeding. Arch Otolaryngol Head Neck Surg 1996; 122: 937–40[Medline]

78 Gallagher JE, Blauth J, Fornadley JA. Perioperative ketorolac tromethamine and postoperative hemorrhage in cases of tonsillectomy and adenoidectomy. Laryngoscope 1995; 105: 606–9[ISI][Medline]

79 Neal BC, Rodgers A, Clark T et al. A systematic survey of 13 randomized trials of non-steroidal anti-inflammatory drugs for the prevention of heterotopic bone formation after major hip surgery. Acta Orthop Scand 2000; 71: 122–8[ISI][Medline]

80 Giannoudis PV, MacDonald DA, Matthews SJ, Smith RM, Furlong AJ, De Boer P. Nonunion of the femoral diaphysis. The influence of reaming and non-steroidal anti-inflammatory drugs. J Bone Joint Surg Br 2000; 82: 655–8[Medline]

81 Murphy PJ, Badia P, Myers BL, Boecker MR, Wright KP Jr. Nonsteroidal anti-inflammatory drugs affect normal sleep patterns in humans. Physiol Behav 1994; 55: 1063–6[ISI][Medline]

82 Murphy PJ, Myers BL, Badia P. Nonsteroidal anti-inflammatory drugs alter body temperature and suppress melatonin in humans. Physiol Behav 1996; 59: 133–9[Medline]

83 Tighe KE, Webb AM, Hobbs GJ. Persistently high plasma morphine-6-glucuronide levels despite decreased hourly patient-controlled analgesia morphine use after single-dose diclofenac: potential for opioid-related toxicity. Anesth Analg 1999; 88: 1137–42[Abstract/Free Full Text]

84 Seideman P, Melander A. Equianalgesic effects of paracetamol and indomethacin in rheumatoid arthritis. Br J Rheumatol 1988; 27: 117–22[ISI][Medline]

85 Seideman P, Samuelson P, Neander G. Naproxen and paracetamol compared with naproxen only in coxarthrosis. Increased effect of the combination in 18 patients. Acta Orthop Scand 1993; 64: 285–8[ISI][Medline]

86 Seideman P. Additive effect of combined naproxen and paracetamol in rheumatoid arthritis. Br J Rheumatol 1993; 32: 1077–82[ISI][Medline]

87 Seideman P. Paracetamol in rheumatoid arthritis. Agents Actions Suppl 1993; 44: 7–12[Medline]

88 McQuay H, Moore A. Estimating relative effectiveness. In: McQuay H, Moore A, eds. An Evidence-Based Resource for Pain Relief. Oxford: Oxford University Press, 1998; 19–31

89 McQuay H, Moore A. Pain measurement, study design, and validity. In: McQuay H, Moore A, eds. An Evidence-Based Resource for Pain Relief. Oxford: Oxford University Press, 1998; 14–8

90 McQuay H, Moore A. Combining data and interpreting the results. In: McQuay H, Moore A, eds. An Evidence-Based Resource for Pain Relief. Oxford: Oxford University Press, 1998; 32–8

91 Hahn TW, Mogensen T, Lund C, Schouenborg L, Rasmussen M. High-dose rectal and oral acetaminophen in postoperative patients-serum and saliva concentrations. Acta Anaesthesiol Scand 2000; 44: 302–6[ISI][Medline]