Ibuprofen and/or paracetamol (acetaminophen) for pain relief after surgical removal of lower wisdom teeth
Abstract
Background
Both paracetamol and ibuprofen are commonly used analgesics for the relief of pain following the surgical removal of lower wisdom teeth (third molars). In 2010, a novel analgesic (marketed as Nuromol) containing both paracetamol and ibuprofen in the same tablet was launched in the United Kingdom, this drug has shown promising results to date and we have chosen to also compare the combined drug with the single drugs using this model. In this review we investigated the optimal doses of both paracetamol and ibuprofen via comparison of both and via comparison with the novel combined drug. We have taken into account the side effect profile of the study drugs. This review will help oral surgeons to decide on which analgesic to prescribe following wisdom tooth removal.
Objectives
To compare the beneficial and harmful effects of paracetamol, ibuprofen and the novel combination of both in a single tablet for pain relief following the surgical removal of lower wisdom teeth, at different doses and administered postoperatively.
Search methods
We searched the Cochrane Oral Health Group'sTrials Register (to 20 May 2013); the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2013, Issue 4); MEDLINE via OVID (1946 to 20 May 2013); EMBASE via OVID (1980 to 20 May 2013) and the metaRegister of Controlled Trials (to 20 May 2013). We checked the bibliographies of relevant clinical trials and review articles for further studies. We wrote to authors of the identified randomised controlled trials (RCTs), and searched personal references in an attempt to identify unpublished or ongoing RCTs. No language restriction was applied to the searches of the electronic databases.
Selection criteria
Only randomised controlled double‐blinded clinical trials were included. Cross‐over studies were included provided there was a wash out period of at least 14 days. There had to be a direct comparison in the trial of two or more of the trial drugs at any dosage. All trials used the third molar pain model.
Data collection and analysis
All trials identified were scanned independently and in duplicate by two review authors, any disagreements were resolved by discussion, or if necessary a third review author was consulted. The proportion of patients with at least 50% pain relief (based on total pain relief (TOTPAR) and summed pain intensity difference (SPID) data) was calculated for all three drugs at both two and six hours postdosing and meta‐analysed for comparison. The proportion of participants using rescue medication over both six and eight hours was also collated and compared. The number of patients experiencing adverse events or the total number of adverse events reported or both were analysed for comparison.
Main results
Seven studies were included, they were all parallel‐group studies, two studies were assessed as at low risk of bias and three at high risk of bias; two were considered to have unclear bias in their methodology. A total of 2241 participants were enrolled in these trials.
Ibuprofen was found to be a superior analgesic to paracetamol at several doses with high quality evidence suggesting that ibuprofen 400 mg is superior to 1000 mg paracetamol based on pain relief (estimated from TOTPAR data) and the use of rescue medication meta‐analyses. The risk ratio for at least 50% pain relief (based on TOTPAR) at six hours was 1.47 (95% confidence interval (CI) 1.28 to 1.69; five trials) favouring 400 mg ibuprofen over 1000 mg paracetamol, and the risk ratio for not using rescue medication (also favouring ibuprofen) was 1.50 (95% CI 1.25 to 1.79; four trials).
The combined drug showed promising results, with a risk ratio for at least 50% of the maximum pain relief over six hours of 1.77 (95% CI 1.32 to 2.39) (paracetamol 1000 mg and ibuprofen 400 mg) (one trial; moderate quality evidence), and risk ratio not using rescue medication 1.60 (95% CI 1.36 to 1.88) (two trials; moderate quality evidence).
The information available regarding adverse events from the studies (including nausea, vomiting, headaches and dizziness) indicated that they were comparable between the treatment groups. However, we could not formally analyse the data as it was not possible to work out how many adverse events there were in total.
Authors' conclusions
There is high quality evidence that ibuprofen is superior to paracetamol at doses of 200 mg to 512 mg and 600 mg to 1000 mg respectively based on pain relief and use of rescue medication data collected at six hours postoperatively. The majority of this evidence (five out of six trials) compared ibuprofen 400 mg with paracetamol 1000 mg, these are the most frequently prescribed doses in clinical practice. The novel combination drug is showing encouraging results based on the outcomes from two trials when compared to the single drugs.
Plain language summary
Ibuprofen versus paracetamol (acetaminophen) for pain relief after surgical removal of lower wisdom teeth
Review question
This review, carried out by the Cochrane Oral Health Group, seeks to compare the effectiveness of two commonly used painkillers, paracetamol and ibuprofen and the combination of both in a single tablet in the relief of pain following surgical removal of lower wisdom teeth.
Background
Worldwide the number of surgical operations to remove wisdom teeth is immense, in England alone approximately 63,000 are removed in National Health Service (NHS) hospitals each year. Many patients need time off work and their quality of life is significantly affected. However, despite these consequences, people are often most concerned about pain following the operation which can be severe. It is suggested that the most intense pain is felt three to five hours after surgery. The pain experienced after oral surgery is widely used as a model to measure the effectiveness of painkillers in general.
Both paracetamol and ibuprofen are commonly used for the relief of pain following the surgical removal of lower wisdom teeth. In 2010, a new painkiller (marketed as Nuromol) containing paracetamol and ibuprofen in the same tablet was licensed for use in the United Kingdom.
All the drugs studied in this review had minimal side effects noted when used correctly for short‐term pain relief.
Study characteristics
The evidence on which this review is based was current as of 20 May 2013. Seven studies with a total of 2241 participants all involving a direct comparison of ibuprofen to paracetamol or the combination of both were included in this review. All participants had surgery to remove a lower wisdom tooth or teeth that required bone removal or at least caused moderate to severe pain. Painkillers were taken after surgery and different doses of the drugs were compared.
The majority of the studies took place in the USA with one in Puerto Rico. Four of the trials took place in clinical research facilities, two in university dental hospitals and one in a private oral surgery clinic. The age of participants differed slightly between studies but was broadly similar, ranging from 15 to 65 years old. All studies included male and female participants.
All the studies included in this review looked only at pain relief and intensity information after a single dose of the painkiller after surgery. It is known that pain does continue after this and the drugs evaluated in this review are normally taken every six to eight hours (maximum of four times per day).
Key results
Ibuprofen is more effective than paracetamol at all doses studied in this review. On limited evidence, the combination of ibuprofen and paracetamol appeared to be no more effective than the single drugs when measured two hours after surgery. However, again on limited evidence, it was found to be more effective than the drugs taken singly when measured at six hours after surgery. Participants taking the combined drug also had a smaller chance of requiring rescue medication.
The information available regarding adverse events from the studies (including nausea, vomiting, headaches and dizziness) indicated that they were comparable between the treatment groups. However, review authors could not formally analyse the data as it was not possible to work out how many adverse events there were in total.
Quality of the evidence
All of the results (outcomes) comparing ibuprofen to paracetamol are of high quality. This means that further research is very unlikely to change our confidence in the estimates of the effect.
When comparing combined versus single drugs, the body of evidence for the proportion of patients with > 50% maximum pain relief (TOTPAR) over two and six hours, was assessed as of moderate quality due to imprecise estimates based on single studies. This means that further research is likely to have an important impact on our confidence in the estimate of the effect. The body of evidence for the use of rescue medication was assessed as being of high quality.
Authors' conclusions
Summary of findings
| Ibuprofen versus paracetamol for pain relief following the surgical removal of lower wisdom teeth | ||||||
| Patient or population: Patients with pain after surgical removal of lower wisdom teeth Control: Paracetamol | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed1 risk | Corresponding risk | |||||
| Control | Ibuprofen | |||||
| Proportion of patients with > 50% maximum pain relief (TOTPAR) over 6 hours ‐ Ibuprofen 400 mg versus paracetamol 1000 mg | Study population | RR 1.47 | 646 | ⊕⊕⊕⊕ | When all doses considered RR = 1.45 (95% CI 1.31 to 1.61) (6 studies, 926 participants; high quality evidence) | |
| 56 per 100 | 83 per 100 | |||||
| Proportion of patients with > 50% maximum pain relief (TOTPAR) over 2 hours ‐ Ibuprofen 400 mg versus paracetamol 1000 mg | Study population | RR 1.30 | 645 | ⊕⊕⊕⊕ | When all doses considered RR = 1.29 (95% CI 1.13 to 1.46) (6 studies, 926 participants; high quality evidence) | |
| 62 per 100 | 81 per 100 | |||||
| Number of patients not using rescue medication at 6 hours (non‐event) Follow‐up: 6 hours | Study population | RR 1.50 | 542 | ⊕⊕⊕⊕ | When all doses considered RR = 1.44 (95% CI 1.13 to 1.64) (5 studies, 823 participants; high quality evidence) | |
| 50 per 100 | 75 per 100 | |||||
| Adverse events | The majority of adverse events were minor in nature and usually included nausea, vomiting, headaches and dizziness. Side effect profiles have not been included in a meta‐analysis as multiple adverse events were recorded in single patients. However, the differences in the observed adverse events for ibuprofen and paracetamol were small | |||||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 This is the median control group risk based on paracetamol being the control group. | ||||||
| Combined (ibuprofen and paracetamol) versus single drugs for pain relief after surgical removal of lower wisdom teeth | ||||||
| Patient or population: Patients with pain after surgical removal of lower wisdom teeth Control: Single drugs | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed1 risk | Corresponding risk | |||||
| Control | Combined (ibuprofen and paracetamol) | |||||
| Proportion of patients with > 50% maximum pain relief (TOTPAR) over 6 hours ‐ Paracetamol 1000 mg/ibuprofen 400 mg versus paracetamol 1000 mg or ibuprofen 400 mg | Study population | RR 1.77 | 170 | ⊕⊕⊕⊝ | ||
| 38 per 100 | 67 per 100 | |||||
| Proportion of patients with > 50% maximum pain relief (TOTPAR) over 2 hours ‐ Paracetamol 1000 mg/ibuprofen 400 mg versus paracetamol 1000 mg or ibuprofen 400 mg | Study population | RR 1.29 | 170 | ⊕⊕⊕⊝ | ||
| 37 per 100 | 48 per 100 | |||||
| Number of patients not using rescue medication at 8 hours ‐ Paracetamol 1000 mg/ibuprofen 400 mg versus paracetamol 1000 mg or ibuprofen 400 mg | Study population | RR 1.60 | 467 | ⊕⊕⊕⊕ | ||
| 50 per 100 | 80 per 100 | |||||
| Adverse events | The majority of adverse events were minor in nature and usually included nausea, vomiting, headaches and dizziness. Side effect profiles have not been included in a meta‐analysis as multiple adverse events were recorded in single patients. However, the differences in the observed adverse events for ibuprofen and paracetamol were small | |||||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 This is the median control group risk based on paracetamol/ibuprofen as single drugs being the control group. 2 Quality of evidence downgraded due to single study and serious imprecision. | ||||||
Background
Description of the condition
In England alone, approximately 63,000 third molars are removed in National Health Service (NHS) hospitals each year (calculated from data available from Hospital Episode Statistics (HES)). Worldwide the number of surgical operations to remove wisdom teeth is immense. Research suggests that wisdom tooth removal has an immediate negative impact on patients' working and social lives: in one study patients took an average of 1.6 days off work, with over one third of patients stating that the surgery had affected their performance at work (Colorado‐Bonnin 2006), and participation in social activities, sports and other hobbies is also negatively affected (Conrad 1999). For many patients quality of life (QoL) is reduced for one to two weeks after surgery (Savin 1997). Postoperative complications may include swelling, bruising and limited mouth opening along with difficulty with eating which can be a major concern to patients and has not been appreciated by healthcare professionals in the past (Ogden 1998). However, patients are often most concerned about postoperative pain, which may be severe. Approximately one in two patients will experience pain despite analgesic therapy, even one week after surgery (Savin 1997). The pain experienced after oral surgery is a validated and widely used pain model for the clinical evaluation of analgesic efficacy (Cooper 1976). Tissue damage produced during surgery releases chemicals that initiate inflammatory pain by activating and sensitising nerve fibre receptors (Loeser 1999). Chemicals include bradykinin, prostaglandins, serotonin and histamine (Dray 1997).
Description of the intervention
Many textbooks of oral surgery practice and drug formularies advocate the use of non‐steroidal anti‐inflammatory drugs (NSAIDs) for the management of postoperative pain, and these drugs have been widely used for pain relief in dentistry for some time (Gobetti 1992). There are now over 50 different NSAIDs on the global market. One of the most commonly prescribed NSAIDs is ibuprofen, with 4.5 million prescriptions for ibuprofen being issued in the United Kingdom (UK) during 2007 (Derry 2009). Ibuprofen has been the subject of much research into its efficacy in postoperative dental pain (Derry 2009). The newer drugs decrease the incidence of gastric perforation, obstruction and bleeding by at least 50% (Boers 2001). However, these drugs provide no cardio‐protection and may be associated with an increased risk of myocardial infarction (Rang 2012).
Ibuprofen has been shown to be an effective analgesic in the control of postoperative dental pain in a number of clinical trials (Hersh 2000; Seymour 1998; Winter 1978). Paracetamol (acetaminophen) has been commercially available since 1953 making it one of the oldest analgesics on the market. Both ibuprofen and paracetamol are amongst the most commonly used analgesics and are widely available without prescription around the world. Paracetamol is of particular value when NSAIDS are contraindicated, perhaps by known hypersensitivity or a history of gastrointestinal ulceration or bleeding (Nguyen 1999). It is also the analgesic of choice to supplement NSAIDS when these alone are expected to be ineffective to control pain (McQuay 1998). Pain intensity following third molar surgery has been suggested to reach its maximum between three to five hours following surgery (Fisher 1988; Seymour 1985) and therefore this pain model is used to test the efficacy of a single analgesic dose.
The combining of analgesic drugs with different modes of action in order to increase the analgesic effect has been well documented (Bromley 2010; Mehlisch 2002). In 2010, a single combination drug containing both paracetamol and ibuprofen was first licensed for use in the UK. In one study using the third molar pain model, the combination analgesic was shown to be a highly effective drug that was comparable with, or superior to, other combination analgesics marketed for severe pain (Daniels 2011). This drug continues to provide encouraging results in analgesic trials. Indeed, a recent Cochrane review (Derry 2013) compared the combined drug (paracetamol plus ibuprofen) with placebo or the same dose of oral ibuprofen alone using the postoperative dental pain model. The authors concluded that ibuprofen plus paracetamol combinations provided better analgesia than either drug alone (at the same dose), with a smaller chance of needing additional analgesia over about eight hours, and with a smaller chance of experiencing an adverse event.
Recent systematic reviews (Collins 1999; Toms 2008) have looked at the efficacy and safety of ibuprofen and paracetamol individually, without direct comparison, for postoperative pain management. These reviews have included the findings of studies involving a wide variety of types of surgery such as inguinal hernia surgery, caesarean section, orthopaedic surgery and including the removal of wisdom teeth. Only one review to date looks at paracetamol specifically in relation to postoperative third molar removal pain (Weil 2007), although two other reviews of single dose postoperative analgesics include subgroup analyses for dental pain only (Moore 2011; Toms 2008). There is some debate as to whether dental pain is different from other pain. It has been suggested that the effect of some analgesics including tramadol was worse for dental pain than for other types of postsurgical pain (Moore 1997).
How the intervention might work
NSAIDs are assumed largely to produce their analgesia as a result of the inhibition of prostaglandin production by the enzyme cyclol‐oxygenase (Malmberg 1992). This prostaglandin inhibition is also responsible for the loss of gastric protection and consequent ulceration and bleeding that can occur. NSAIDs have the ability to inhibit the fatty acid cyclo‐oxygenase enzyme, thereby inhibiting the production of prostaglandins and thromboxanes (Rang 2012). Cyclo‐oxygenases (COX) oxidise arachidonate producing unstable intermediate prostaglandins PGG2 and PGH2. There are two main isoforms of COX: COX‐1 is a constitutive enzyme and COX‐2 is often induced by inflammatory stimuli. Prostaglandins and thromboxanes are nociceptive initiators which lead to painful sensations in the body. It is thought that the anti‐inflammatory, analgesic and antipyretic actions of the NSAIDs are related to inhibition of COX‐2 and that the unwanted effects, particularly the gastrointestinal consequences are related to inhibition of COX‐1. A recent development has been the synthesis of selective COX‐2 inhibitors; examples include celecoxib, etoricoxib and lumiracoxib (Neal 2012).
Paracetamol (acetaminophen) is a non‐opioid analgesic possessing antipyretic activity and is effective in relieving pain with a low incidence of adverse effects, it has proven to be a safe, effective drug for the treatment of postoperative pain following the surgical removal of lower wisdom teeth (Weil 2007). Paracetamol is often grouped with the NSAID family, however, it is considered only to have relatively weak anti‐inflammatory activity (Rang 2012). Although the mechanism of action was not fully understood until recently, it is now thought that paracetamol is a selective inhibitor of the newly described COX‐3 enzyme, a cyclo‐oxygenase‐1 variant, in the central nervous system where it acts as a prodrug. It is deacylated to p‐aminophenol and in turn conjugated with arachidonic acid to form N‐arachidonoyl‐phenolamine. This compound is an endogenous cannabinoid, acting on CB1 receptors, and is also an agonist at TRPV1 receptors (Bromley 2010). This inhibition could represent a primary central mechanism by which paracetamol decreases pain and possibly fever (Chandrasekharan 2002). It also has been shown to be an effective analgesic in the control of postoperative dental pain in a number of clinical trials (Bentley 1987; Kiersch 1994; Mehlisch 1990).
Why it is important to do this review
In this review we investigated the optimal dose of ibuprofen versus paracetamol by direct comparison, taking into account the side effects of different doses of the drugs. This would inform dentists, oral surgeons and their patients of the best strategy for best pain relief when considering ibuprofen or paracetamol (or a combination of both) following the surgical removal of wisdom teeth.
Objectives
-
To discover which analgesic has the best efficacy for managing postoperative pain using the third molar model.
-
To assess the efficacy of novel combination drugs including both agents in the same tablet and to compare this to the individual drugs being administered at the same time.
-
To assess the harmful effects of ibuprofen and paracetamol, and the combination drugs at different doses administered postoperatively.
Methods
Criteria for considering studies for this review
Types of studies
All randomised controlled double‐blinded clinical trials. Cross‐over studies were included provided there was a wash out period of at least 14 days.
Types of participants
Patients of all health states without intolerances/allergies to the study drugs who required the surgical removal of a lower wisdom tooth or teeth that required bone removal or at least having a baseline pain intensity of moderate to severe pain. Patients who required removal of an additional tooth or teeth were also included. Surgery was undertaken under local anaesthesia, intravenous sedation or general anaesthesia. Patients taking concurrent analgesia were excluded.
Types of interventions
Ibuprofen, paracetamol or a combination of both given as a single dose postoperatively by mouth in any dose and in any formulation (for example, immediate or slow release).
Types of outcome measures
-
Pain relief (visual analogue scale (VAS), categorical verbal rating, verbal numerical scale, global subjective efficacy ratings and other categorical rating scales) and derived pain relief outcomes extracted were TOTPAR (total pain relief), and SPID (summed pain intensity difference) over two to six hours (dichotomous).
-
Side effects (for example, gastrointestinal, hepatic and renal) (binary).
-
Use of rescue medication within six to eight hours of single dose analgesic administration.
Search methods for identification of studies
For the identification of studies included or considered for this review, we developed detailed search strategies for each database searched. These were based on the search strategy developed for MEDLINE (OVID) but revised appropriately for each database (Appendix 3). The search strategy used a combination of controlled vocabulary and free text terms and was linked with the Cochrane Highly Sensitive Search Strategy (CHSSS) for identifying randomised trials (RCTs) in MEDLINE: sensitivity maximising version (2008 revision) as referenced in Chapter 6.4.11.1 and detailed in box 6.4.c of the Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011) (Higgins 2011). The search of EMBASE was linked to the Cochrane Oral Health Group filter for identifying RCTs.
Electronic searches
We searched the following electronic databases:
-
the Cochrane Oral Health Group's Trials Register (to 20 May 2013) (Appendix 1);
-
the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochane Library 2013, Issue 4) (Appendix 2);
-
MEDLINE (1946 to 20 May 2013) (Appendix 3);
-
EMBASE (1980 to 20 May 2013) (Appendix 4);
-
metaRegister of Controlled Trials (www.controlled‐trials.com) (to 20 May 2013) (Appendix 5).
There were no language restrictions applied in the searches of the electronic databases. Where necessary, translations into English were obtained.
Searching other resources
Unpublished studies
Authors of RCTs identified were contacted in order to obtain further information about the trial and to attempt to identify unpublished or ongoing studies. We also wrote to manufacturers of analgesic pharmaceuticals.
Handsearching
Only handsearching done as part of the Cochrane Worldwide Handsearching Programme and uploaded to CENTRAL was included (see the Cochrane Masterlist for details of journal issues searched to date).
The bibliographies of papers and review articles were checked for further studies. Personal references were also searched.
Data collection and analysis
Selection of studies
Two review authors independently and in duplicate scanned the titles and abstracts (when available) of all reports identified. For studies appearing to meet the inclusion criteria, or for which there were insufficient data in the title and abstract to make a clear decision, the full report was obtained and assessed independently and in duplicate by two review authors to establish whether the studies met the inclusion criteria or not. Disagreements were resolved by discussion. Where resolution was not possible, a third review author was consulted. All studies meeting the inclusion criteria underwent a risk of bias assessment and data extraction. Studies rejected at this or subsequent stages were recorded in the table of excluded studies, and reasons for exclusion recorded.
Data extraction and management
Two review authors independently and in duplicate extracted data using specially designed data extraction forms. Any disagreement was discussed and a third review author consulted where necessary. Authors were contacted for clarification of missing information. Data were excluded until further clarification was available if agreement was not reached.
For each trial the following data were recorded.
-
Year of publication, country of origin, setting and source of study funding.
-
Details of the participants including demographic characteristics and criteria for inclusion.
-
Details on the study design (parallel group or cross‐over design).
-
Details on the type of intervention.
-
Details of the outcomes reported, including method of assessment and time intervals.
-
Details of withdrawals and drop‐outs by study group.
-
Details of side effects and adverse events.
Assessment of risk of bias in included studies
The domains considered were: random sequence generation (selection bias), allocation concealment (selection bias), blinding (performance and detection bias), incomplete outcome data (attrition bias), and selective reporting (reporting bias) along with any other bias thought to be relevant by the authors.
In summary, risk of bias in the included analgesic studies was assessed by the following criteria.
| Bias | Criteria for low risk of bias in included studies |
| Random sequence generation (selection bias) | Method of randomisation must be clearly stated |
| Allocation concealment (selection bias) | Method of concealment of allocation to interventions must be clearly stated |
| Blinding (performance and detection bias) | Method of blinding of all parties involved in the study must be detailed |
| Incomplete outcome data (attrition bias) | All primary outcome measures must be reported on as detailed in the method |
| Selective reporting (reporting bias) | All drop‐outs and missing data must be accounted for, adverse events must be included in the analysis |
| Other bias | Method of anaesthetic given is clearly defined and unlikely to be a cause of bias in the trial |
Measures of treatment effect
Authors commonly report on the results of analgesic trials using mean data with associated standard deviations, this is a problem as the data may be asymmetrically distributed and if used in meta‐analyses will lead to potentially erroneous conclusions (Moore 1997a). It is therefore important to derive dichotomous data from the continuous data presented in trials prior to using the data in meta‐analyses. The team at the Oxford Pain Relief Unit and Nuffield Department of Anaesthetics have derived a method for dichotomising these data; the detailed background and verification were published over three papers (Moore 1996; Moore 1997a; Moore 1997b).
From the data presented in the trials, the proportion of patients achieving 50% pain relief (50% maximum TOTPAR) was calculated and used in the meta‐analysis. Other Cochrane reviews have made use of these measures in their analyses (Derry 2009; Toms 2008; Weil 2007). SPID essentially measures the same thing as TOTPAR. If data were unavailable to calculate TOTPAR, SPID would have been calculated. If data on both TOTPAR and SPID were available, TOTPAR was chosen in preference. Outcomes were assessed for two hours and six hours postdosing (where possible). For these dichotomous outcomes, the estimate of an intervention was expressed as risk ratios together with 95% confidence intervals.
Unit of analysis issues
The unit of analysis is individual patients, although appropriate cross‐over studies were included. If data from patients in the same treatment group were used in more than one dose comparison for meta‐analysis, the number of patients was split between the groups. For example, if 50 out of 100 patients achieved the desired outcome by taking 1000 mg of paracetamol and these data were to be compared with two different doses of ibuprofen, the figures for the analysis were halved (25 out of 50 for each comparison).
Dealing with missing data
As described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), there are several types of missing data in a systematic review or meta‐analysis. The problem of missing studies and outcomes is addressed in the Assessment of reporting biases part of this review. A common problem is missing summary data, such as standard deviations for continuous outcomes, or separate sample sizes for each intervention group. Missing summary data were not a reason to exclude a study from the review and methods outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) were used for imputing missing standard deviations. In the analysis we made the assumption that the data were missing at random, so we included only available data. The authors were contacted where possible for missing data. Cross‐over studies data would be meta‐analysed according to the methods outlined in Elbourne 2002.
Assessment of heterogeneity
Prior to meta‐analysis, studies were assessed for clinical homogeneity with respect to type of therapy, control group and the outcomes. Clinically heterogeneous studies were not combined in a meta‐analysis, but described in a narrative way. For studies judged as clinically homogeneous, statistical heterogeneity was tested by Q test (Chi2) and the I2 statistic. We interpreted a Chi2 test resulting in a P value < 0.10 as indicating statistically significant heterogeneity. In order to assess and quantify the possible magnitude of inconsistency (i.e. heterogeneity) across studies, we used the I2 statistic with a rough guide for interpretation as follows: 0% to 40% might not be important; 30% to 60% may represent moderate heterogeneity; 50% to 90% may represent substantial heterogeneity; and 75% to 100% considerable heterogeneity.
Assessment of reporting biases
Possible reporting biases were assessed on two levels: within‐study and between‐study.
Within‐study selective outcome reporting was examined as part of the overall risk of bias assessment (Assessment of risk of bias in included studies). Outcomes listed in the methods sections on a publication were compared against those whose results were reported. Where some indications of reporting bias were found, study authors were contacted for clarification.
If there were least 10 studies included in a meta‐analysis in the review, a funnel plot of effect estimates against their standard errors was planned to assess a possible between‐study reporting bias. If an asymmetry of the funnel plot was found by inspection and confirmed by statistical tests, possible explanations would be considered and taken into account in the interpretation of the overall estimate of treatment effects.
Data synthesis
Meta‐analysis was conducted only for studies with similar comparisons reporting the same outcome measures. Risk ratios were used to combine dichotomous data, and weighted mean differences for continuous data (if data had been available), using random‐effects models provided there were more than three studies eligible for meta‐analysis. Different dose comparisons were presented as subgroups and we divided up the numbers of patients between subgroups to avoid 'double counting'.
Subgroup analysis and investigation of heterogeneity
Subgroup analysis was conducted for.
-
Subgroups were used for different dose comparisons.
-
Where different types of formulation of ibuprofen or paracetamol were used: for instance, immediate release versus slow release.
Had data allowed, the following subgroup analyses would have been undertaken:
Where patients had undergone surgery with local anaesthesia alone, local anaesthesia and intravenous sedation, general anaesthesia alone and general anaesthesia with local anaesthetic.
Sensitivity analysis
Primary meta‐analyses included all studies irrespective of their risk of bias. Sensitivity analysis was planned to assess how the results of meta‐analysis were affected if studies at high risk of bias were excluded from the analysis. A sensitivity analysis was also planned to take into account the sources of funding of the included studies.
Results
Description of studies
Seven studies were included, they were all parallel‐group studies (Characteristics of included studies).
Results of the search
1539 studies were identified through the searching. Initial screening identified 13 potential articles for which full‐text was retrieved (Figure 1). After closer examination, six were excluded (Characteristics of excluded studies), leaving seven included studies (Characteristics of included studies).
Study flow diagram.
Included studies
Seven studies were included in this review. The seven studies contained data on 2241 participants. All of these studies included a direct comparison of ibuprofen to paracetamol or the combination of both agents in the same drug (along with other analgesics in some trials: data not used in this review) in the postoperative third molar surgery pain model.
Characteristics of the trial setting and investigators
The majority of the trials (six) were conducted in the USA, with one trial conducted in Puerto Rico (Olson 2001). Three of the trials were conducted by the same lead author, albeit with different collaborators (Mehlisch 1995; Mehlisch 2010; Mehlisch 2010a). Four of the trials were completed in clinical research facilities (Daniels 2009; Mehlisch 1995; Mehlisch 2010; Mehlisch 2010a), two in university dental hospitals (Hersh 2000; Olson 2001) and one in a private oral surgery clinic (Forbes 1990).
Characteristics of the participants
The participants were broadly similar in the included trials, all contained the following exclusion criteria:
-
history of significant disease
-
ongoing painful conditions (other than the third molar(s) scheduled for removal)
-
allergy/intolerance to the study drugs
-
patients currently taking long‐term analgesics
-
malabsorption states (not mentioned in Mehlisch 1995)
-
gastrointestinal complaints (not mentioned in Mehlisch 1995)
-
psychotic illness or drug abuse (not mentioned in Mehlisch 1995)
-
concomitant medication that would interfere with the study drugs (not mentioned in Forbes 1990)
-
pregnancy and/or breastfeeding (not mentioned in Mehlisch 1995)
-
migraine (not mentioned in Forbes 1990; Hersh 2000 or Olson 2001).
The age range of the participants was slightly different across the studies but broadly similar. All studies included both male and female participants. In Forbes 1990 and Mehlisch 1995, the age range was > 15 years, in Hersh 2000 and Mehlisch 2010a, it was > 16 years. Age ranges applied in Daniels 2009 and Mehlisch 2010 (16 to 40 years), with a range of 16 to 65 years used by Olson 2001.
Characteristics of the interventions
All seven studies compared paracetamol with ibuprofen and two compared the combination drugs with their individual constituents (Mehlisch 2010; Mehlisch 2010a) (Characteristics of included studies; Additional Table 1). The studies included data on the following doses of analgesics.
|
| Daniels 2009 | Forbes 1990 | Hersch 2000* | Mehlisch 1995 | Mehlisch 2010 | Mehlisch 2010a | Olson 2001* |
| Paracetamol 500 mg
|
|
|
|
|
| √ |
|
| Paracetamol 600 mg
|
| √ |
|
|
|
|
|
| Paracetamol 1000 mg
| √ |
| √ | √ | √ | √ | √ |
| Ibuprofen 200 mg
|
|
| √ |
|
| √ |
|
| Ibuprofen 400 mg
| √ | √ | √ | √ | √ | √ | √ |
| Ibuprofen 512 mg
| √ |
|
|
|
|
|
|
| Paracetamol 250 mg/ibuprofen 100 mg
|
|
|
|
|
| √ |
|
| Paracetamol 500 mg/ibuprofen 200 mg
|
|
|
|
| √ | √ |
|
| Paracetamol 1000 mg/ibuprofen 400 mg
|
|
|
|
| √ | √ |
|
*Liquigel formula.
-
Paracetamol 500 mg (Mehlisch 2010a).
-
Paracetamol 600 mg (Forbes 1990).
-
Paracetamol 1000 mg (Daniels 2009; Hersh 2000; Mehlisch 1995; Mehlisch 2010; Mehlisch 2010a; Olson 2001).
-
Ibuprofen 200 mg Hersh 2000; Mehlisch 2010a).
-
Ibuprofen 400 mg: all of the included studies had data on this drug (Daniels 2009; Forbes 1990; Hersh 2000; Mehlisch 1995; Mehlisch 2010; Mehlisch 2010a; Olson 2001).
-
Ibuprofen 512 mg liquigel formula (Daniels 2009).
-
Paracetamol 250 mg/ibuprofen 100 mg combined drug (Mehlisch 2010a).
-
Paracetamol 1000 mg/ibuprofen 400 mg combined drug (Mehlisch 2010; Mehlisch 2010a).
-
Paracetamol 500 mg/ibuprofen 200 mg combined drug (Mehlisch 2010; Mehlisch 2010a).
All of the studies were double‐blinded, parallel‐group randomised controlled trials with dummy medications being issued in Daniels 2009; Mehlisch 1995 and Olson 2001. All of the studies provided rescue medication (Additional Table 2), the drug(s) provided were not detailed in Mehlisch 1995 and Olson 2001 and included a variety of different drugs, with some being administered intramuscularly (Characteristics of included studies).
| Study | Use of rescue medication (RM) | ||||
| Mean time to RM | Use of RM (%) | Use of RM (n) | Total | Observation period | |
| Daniels 2009 | 6 hours | ||||
| Paracetamol 1000 | 44 | 35 | 80 | ||
| Ibuprofen 400 | 23 | 18 | 80 | ||
| Ibuprofen 512 | 33 | 26 | 80 | ||
| Forbes 1990 | 6 hours | ||||
| Paracetamol 600 | 3.89 hours | 81 | 29 | 36 | |
| Ibuprofen 400 | 4.63 hours | 63 | 20 | 32 | |
| Hersch 2000 | 6 hours | ||||
| Paracetamol 1000 | 51 | 32 | 63 | ||
| Ibuprofen 200 | 31 | 19 | 61 | ||
| Ibuprofen 400 | 22 | 13 | 59 | ||
| Mehlisch 1995 | 6 hours | ||||
| Paracetamol 1000 | 60 | 60 | 101 | ||
| Ibuprofen 400 | 26 | 25 | 98 | ||
| Mehlisch 2010 | 8 hours | ||||
| Paracetamol 1000 | 261 minutes | 71 | 24 | 34 | |
| Ibuprofen 400 | 296 minutes | 68 | 47 | 69 | |
| Paracetamol 1000/ibuprofen 400 | 376 minutes | 31 | 21 | 67 | |
| Paracetamol 500/ibuprofen 200 | 329 minutes | 61 | 20 | 33 | |
| Mehlisch 2010a | 8 hours in phase 1 | ||||
| Paracetamol 500 | 74 | 56 | 76 | ||
| Paracetamol 1000 | 69 | 51 | 74 | ||
| Ibuprofen 200 | 39 | 29 | 75 | ||
| Ibuprofen 400 | 28 | 21 | 74 | ||
| Paracetamol 250/ibuprofen 100 | 38 | 27 | 71 | ||
| Paracetamol 1000/ibuprofen 400 | 28 | 42 | 149 | ||
| Paracetamol 500/ibuprofen 200 | 22 | 31 | 143 | ||
| Olson 2001 | 6 hours | ||||
| Paracetamol 1000 | 38 | 25 | 66 | ||
| Ibuprofen 400 | 21 | 14 | 67 | ||
Use of rescue medication
Rescue medication was provided in all studies, all studies contained data on the percentage of patients taking rescue medication over the study period which was six hours in all of the studies with the exception of Mehlisch 2010 and Mehlisch 2010a which had eight‐hour periods of assessment.
Number of third molars removed
In Daniels 2009; Hersh 2000 and Olson 2001, at least one third molar impacted in bone was removed, in Forbes 1990 it states that at least one third molar was removed but it does not state whether bone removal was carried out. In Mehlisch 1995, at least two third molars were removed, one of which was impacted in bone. In the later Mehlisch studies (Mehlisch 2010; Mehlisch 2010a), the participants had three or four third molars removed, two of which had to be impacted in bone in the mandible. It is thought that the removal of bone causes severe pain following the removal of third molars (Coulthard 2009).
Type of anaesthetic used
The anaesthetic used for the surgical procedure varied in the studies. In two studies, general anaesthetic with supplemental local anaesthetic was used (Forbes 1990; Mehlisch 1995). Local anaesthetic alone was used in one study (Olson 2001). Local anaesthetic with supplemental sedation was used in four studies, one using inhalation sedation with nitrous oxides (Daniels 2009), two with nitrous oxide, diazepam and a barbituate (Mehlisch 2010; Mehlisch 2010a) and in one study (Hersh 2000) "most patients" received intravenous conscious sedation. It is important to be aware of whether trial participants were sedated as certain sedative agents commonly used in oral surgery, namely midazolam, have been shown to have an analgesic effect (Coulthard 1992; Coulthard 1993). This effect could have influenced the results of the trials.
Number of doses of analgesic given
For the purposes of data extraction in this review, only the data from the first postoperative dose were used. In Daniels 2009; Hersh 2000; Mehlisch 1995; Mehlisch 2010 and Olson 2001 the study period included only the data for the first six to eight hours following the first dose of analgesic. In Mehlisch 2010a, three doses were provided and in Forbes 1990 there were 15. None of the included studies provided preoperative analgesics.
Baseline pain intensity
It is important that baseline pain intensity measures are included in trials; baseline pain provides the reference point from which the degree of pain relief or an increase or decrease in pain intensity can be measured. In analgesic trials, it is necessary for participants to be experiencing a certain amount of pain before the analgesic is administered in order for the efficacy of the drug to be tested. If participants have no pain at the outset then there is no point in providing analgesics as there will be no improvement in pain scores. The included studies varied in their criteria for baseline pain intensity (as in how much pain a participant has to be experiencing prior to receiving a dose of the test analgesic). Five of the studies used a visual analogue scale (VAS) with a dosing threshold of > 50 mm (Daniels 2009; Hersh 2000; Mehlisch 2010; Mehlisch 2010a; Olson 2001); in the Daniels 2009 study, the VAS had to be between 50 mm and 85 mm prior to dosing, the reason for defining an upper limit for VAS was not specified. In the other two trials (Forbes 1990 and Mehlisch 1995), the dosing threshold was described using a categorical scale whereby the participant had to state that they were suffering from moderate or severe pain prior to dosing (this roughly translates to the VAS threshold used by the more recent studies).
Excluded studies
Refer to Characteristics of excluded studies for details of the excluded studies. In total, six studies were excluded following a thorough read through by two of the review authors. Bjornsson 2003 and Chopra 2009 were excluded due to being multiple dose studies, we were also unable to extract reliable single dose data from these trials. Dionne 1983; Merry 2010 and Ozkan 2010 were also excluded for these reasons and the use of pre‐emptive analgesia in the trials, it was felt that this would introduce an unacceptable level of bias to the review if included. Ikeda 2002 was excluded due to there being no published paper with the results from the trial available, only an abstract from a conference.
Risk of bias in included studies
See Figure 2 and Figure 3 for a summary of risk of bias in the included trials. Two studies were shown to be at low risk of bias across all of the domains (Mehlisch 2010; Mehlisch 2010a). Three studies were assessed as at high risk of bias overall (Forbes 1990; Hersh 2000;Mehlisch 1995) and two were considered to have unclear bias in their methodology (Daniels 2009; Olson 2001).
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
Five out of the seven studies reported the random sequence and were assessed at low risk, Olson 2001 was shown to have unclear risk of bias in this area due to the sponsor being responsible for the allocation of interventions, and Hersh 2000 was found to be at high risk where no detail was given as to how the patients were randomised.
Mehlisch 2010; Mehlisch 2010a and Olson 2001 reported on allocation concealment and were judged to be at low risk of bias. Forbes 1990; Hersh 2000 and Mehlisch 1995 did not give any information on allocation concealment and were found to be at high risk of bias. In Daniels 2009 study medication appeared to have been matched with placebo, however, the study was judged to be at unclear risk of bias because it was not clear whether the allocation concealment was adequate.
In analgesic trials it is important that the dosing sequence is blinded to the clinicians, pharmacists, nurses and participants. The randomisation sequence is usually kept by a third party and only broken if patient safety is at risk.
Blinding
Forbes 1990; Mehlisch 2010 and Mehlisch 2010a were judged to be at low risk of bias. Daniels 2009; Hersh 2000; Mehlisch 1995 and Olson 2001 were found to be at unclear risk of bias as they claimed to be 'double blind' but did not state explicitly how the blinding process was performed.
Incomplete outcome data
All of the studies were found to be at low risk with the exception of Mehlisch 1995, which was found to be at unclear risk of bias due to one patient not completing any analysis and therefore being excluded from data collection. The reasons for this were not fully explored in the paper.
It is important to note that in analgesic trials, the majority of data is gained from participant‐reported outcomes, this therefore limits the influence the investigators can have over the results.
Selective reporting
All seven of the trials were found to be at low risk of bias in this domain. All adverse effects and intended outcomes were reported on.
Other potential sources of bias
Five trials were found to be at low risk of bias with two (Hersh 2000; Mehlisch 1995) being judged as at high risk of other potential sources of bias. In Hersh 2000, this was due to the use of sedation for some patients and not others, with no indication as to how the decision was reached or the randomisation involved in the choice of anaesthetic. This was thought to introduce a potential bias in the results. In Mehlisch 1995, no detail was given on how the study medications were distributed and it was not clear as to whether the medications would be identifiable to the study participants or the assessors.
In analgesic trials, it is important that the participants do not know which medication they are taking, this is usually achieved by producing identical packaging/tablet size for all of the potential dosing regimens. Three of the included studies also used dummy medication (Daniels 2009; Mehlisch 1995; Olson 2001), so that the participants would all have an identical number of pills to take regardless of their dosing regimen.
Effects of interventions
See: Summary of findings for the main comparison Ibuprofen versus paracetamol for pain relief following the surgical removal of lower wisdom teeth; Summary of findings 2 Combined (ibuprofen and paracetamol) versus single drugs for pain relief after surgical removal of lower wisdom teeth
Comparison 1: Ibuprofen versus paracetamol
Outcome TOTPAR (total pain relief) ‐ greater than 50% pain relief over six hours
This comparison comprised four subgroups with different doses of ibuprofen and paracetamol, including six trials (n = 926) (Daniels 2009; Forbes 1990; Hersh 2000; Mehlisch 1995; Mehlisch 2010; Olson 2001) (Analysis 1.1). There was no difference between the subgroups (P value = 0.53), and the overall risk ratio was 1.45 (95% confidence interval (CI) 1.31 to 1.61; P value < 0.00001), indicating that 45% more patients achieved at least 50% of the maximum pain relief over six hours in the ibuprofen group (with doses between 200 mg and 512 mg) compared to the paracetamol group (doses 600 mg and 1000 mg). There was no evidence of any heterogeneity (P value = 0.41; I2 = 3%).
The most frequently assessed dose was ibuprofen 400 mg compared with paracetamol 1000 mg (five trials; n = 646) (Daniels 2009; Hersh 2000; Mehlisch 1995; Mehlisch 2010; Olson 2001). The pooled risk ratio was 1.47 (95% CI 1.28 to 1.69; P value < 0.00001), indicating 47% more patients achieved at least 50% of the maximum pain relief over six hours in the ibuprofen group. There was no evidence of statistical heterogeneity (P value = 0.30; I² = 19%).
Outcome TOTPAR ‐ greater than 50% pain relief over two hours
This comparison comprised four subgroups with different doses of ibuprofen and paracetamol, including six trials (Daniels 2009; Forbes 1990; Hersh 2000; Mehlisch 1995; Mehlisch 2010; Olson 2001) (Analysis 1.2). There was no difference between the subgroups (P value = 0.48), and the overall risk ratio was 1.29 (95% CI 1.13 to 1.46; P value < 0.00001), indicating that 29% more patients achieved at least 50% of the maximum pain relief over two hours in the ibuprofen group (doses 200 mg to 512 mg) compared to the paracetamol group (doses 600 mg and 1000 mg). There was little evidence of any heterogeneity (P value = 0.13; I2 = 38%).
Again, the most frequently assessed comparison was ibuprofen 400 mg compared with paracetamol 1000 mg (five trials; n = 645) (Daniels 2009; Hersh 2000; Mehlisch 1995; Mehlisch 2010; Olson 2001). The pooled risk ratio was 1.30 (95% CI 1.09 to 1.55; P value = 0.003), indicating 30% more patients achieved at least 50% of the maximum pain relief over two hours in the ibuprofen group. There was evidence of statistical heterogeneity (P value = 0.08; I2 = 52%).
Number of patients using rescue medication at six hours
This comparison comprised four subgroups with different doses of ibuprofen and paracetamol, including five trials (Daniels 2009; Forbes 1990; Hersh 2000; Mehlisch 1995; Olson 2001) (Analysis 1.3). There was no difference between the subgroups (P value = 0.49), and the overall risk ratio was 1.44 (95% 1.26 to 1.64; P value < 0.00001), indicating that 44% fewer patients used rescue medication over six hours in the ibuprofen group (doses 200 mg to 512 mg) compared to the paracetamol group (doses 600 mg and 1000 mg). There was no evidence of any heterogeneity (P value = 0.30; I2 = 16%).
Number of patients using rescue medication at eight hours
This comparison comprised four subgroups with different doses of ibuprofen and paracetamol, including two trials (Mehlisch 2010; Mehlisch 2010a) (Analysis 1.4). There was no difference between the subgroups (P value = 0.48), and the overall risk ratio was 2.02 (95% CI 1.57 to 2.60; P value < 0.00001), indicating that twice as many patients used rescue medication in the paracetamol groups (doses 600 mg and 1000 mg) than did in the ibuprofen groups (doses 200 mg to 512 mg) over an eight‐hour period. There was possible evidence of some heterogeneity (P value = 0.22; I2 = 30%).
Comparison 2: Combined (ibuprofen and paracetamol) versus single drugs
The four outcomes for TOTPAR were based on data from one trial Mehlisch 2010, therefore they cannot be considered as meta‐analyses. All of the comparisons were between paracetamol 1000 mg and ibuprofen 400 mg in the same tablet, and the same constituent drugs given as single tablets. It was not possible to derive TOTPAR and/or SPID (summed pain intensity difference) data from Mehlisch 2010a as the trial used a two‐stage design, despite contact with the authors, we did not obtain the specific data required to dichotomise the trial results for meta‐analysis.
Outcome TOTPAR ‐ greater than 50% pain relief over six hours
This comparison demonstrates a risk ratio of 1.77 (95% CI 1.32 to 2.39; P value = 0.0002), indicating that 77% more patients achieved at least 50% of the maximum pain relief over six hours in the combined drug group as did in the single drug group (paracetamol 1000 mg and ibuprofen 400 mg) (Analysis 2.1).
Outcome TOTPAR ‐ greater than 50% pain relief over two hours
The results for TOTPAR at two hours showed a similar preference for the combined drug formulation over the single drugs with TOTPAR demonstrating a risk ratio of 1.29 (95% CI 0.91 to 1.85; P value = 0.15), indicating that 29% more patients achieved maximum pain relief over two hours in the combined drug group as did in the single drug group (paracetamol 1000 mg and ibuprofen 400 mg). The confidence interval is wide so therefore there is no evidence to suggest that the combined drug is any better or worse than the single drugs (Analysis 2.2).
Number of patients using rescue medication at eight hours ‐ Combined drug versus single drugs
The results of two studies were analysed (Mehlisch 2010; Mehlisch 2010a) (Analysis 2.3) comparing the efficacy of a combination of paracetamol 1000 mg/500 mg with ibuprofen 400 mg/200 mg in the same pill with the individual constituent drugs taken together. The overall risk ratio was 1.60 (95% CI 1.36 to 1.88; P value < 0.00001) indicating that 60% more patients used rescue medication over six hours in the individual constituent drug group compared to the combined drug group. There was evidence of substantial heterogeneity (P value = 0.02; I2 = 82%), however this meta‐analysis included only data from two trials and has to be interpreted with caution.
In all of the above comparisons, the combined formula of the drug was favoured (see forest plots).
Side effects profile
All studies had information on the adverse events observed during the entire study period. This information is used to create the safety profile of the study drugs. Data on serious or severe adverse events were also collected and collated as a percentage of total adverse events (Additional Table 3). The vast majority of adverse events were minor in nature and usually included nausea, vomiting, headaches and dizziness. No severe adverse events were thought to be definitely linked to the analgesic drugs or placebos used. It is worth noting that these data were collected in the immediate postoperative period following surgery under local anaesthetic with additional sedation or general anaesthetic in most cases (all studies except Olson 2001 which used local anaesthetic alone), the anaesthetic drugs could be related to the adverse events observed. In Additional Table 3, there is evidence that the frequency of observed adverse events is slightly lower in the Olson 2001 study, adding further weight to this argument. In Mehlisch 2010, the observed frequency of adverse events was high, the authors explained that the events were likely to have been caused by the heavy sedation used for surgery.
| Study | Adverse events (AE) | |||
| Total AE (n) | Total AE % | Serious/Severe AE | Serious/Severe AE % | |
| Daniels 2009 | ||||
| Paracetamol 1000 | 25 (81) | 31 | 3 | 12 |
| Ibuprofen 400 | 19 (80) | 24 | 1 | 3 |
| Ibuprofen 512 | 24 (80) | 30 | 2 | 6 |
| Forbes 1990 | ||||
| Paracetamol 600 | 5 (41) | 12 | Not specified | Not specified |
| Ibuprofen 400 | 8 (43) | 19 | Not specified | Not specified |
| Hersch 2000 | ||||
| Paracetamol 1000 | 12 (63) | 19 | 0 | 0 |
| Ibuprofen 200 | 7 (61) | 11 | 0 | 0 |
| Ibuprofen 400 | 4 (59) | 7 | 0 | 0 |
| Mehlisch 1995 | ||||
| Paracetamol 1000 | 17 (101) | 17 | 0 | 0 |
| Ibuprofen 400 | 12 (98) | 12 | 0 | 0 |
| Mehlisch 2010 | ||||
| Paracetamol 1000 | 24 (34) | 71 | 11 | 46 |
| Ibuprofen 400 | 39 (69) | 57 | 14 | 36 |
| Paracetamol 1000/ibuprofen 400 | 38 (67) | 57 | 11 | 29 |
| Paracetamol 500/ibuprofen 200 | 14 (33) | 42 | 6 | 43 |
| Mehlisch 2010a | ||||
| Paracetamol 500 | 38 (76) | 50 | 7 | 18 |
| Paracetamol 1000 | 30 (74) | 40 | 5 | 17 |
| Ibuprofen 200 | 31 (75) | 41 | 7 | 23 |
| Ibuprofen 400 | 33 (74) | 45 | 5 | 15 |
| Paracetamol 250/ibuprofen 100 | 22 (71) | 31 | 8 | 36 |
| Paracetamol 1000/ibuprofen 400 | 44 (149) | 30 | 15 | 34 |
| Paracetamol 500/ibuprofen 200 | 51 (143) | 36 | 7 | 14 |
| Olson 2001 | ||||
| Paracetamol 1000 | 10 (66) | 15 | 1 | 10 |
| Ibuprofen 400 | 7 (67) | 11 | 2 | 29 |
Side effect profiles have not been included in a meta‐analysis as multiple adverse events were recorded in single patients, it was not possible from the data to work out how many adverse events there were in total. However, Additional Table 3 shows that the differences in the observed adverse events for ibuprofen and paracetamol were small and there were no apparent differences between the groups.
Summary of findings
Using the software GRADEprofiler 3.6, the quality of the body of evidence was assessed for both comparisons: ibuprofen versus paracetamol, and combined (ibuprofen and paracetamol) versus single drugs. TOTPAR, and use of rescue medication were assessed as SPID is measuring the same thing as TOTPAR. A summary of these findings for the two comparisons is shown in summary of findings Table for the main comparison and summary of findings Table 2. These tables show that all of the outcomes for comparing ibuprofen versus paracetamol are assessed as at high quality. The comparisons for combined (ibuprofen and paracetamol) versus single drugs for TOTPAR are both moderate as they were downgraded due to being based on single studies and the imprecision of estimate. The evidence for use of rescue medication was assessed as high quality.
Discussion
The surgical removal of lower third molar teeth continues to be a frequently performed surgical procedure carried out by oral surgeons worldwide. Adequate management of pain is of paramount importance to both the patient and surgeon, and also those involved in the commissioning of services as pain experience can be used as a method for measuring the quality of an oral surgey service. All healthcare decisions should be supported by a sound evidence base, this review will form part of the evidence base that oral surgeons should look to when making decisions on how best to manage their patient's postoperative pain.
Summary of main results
Ibuprofen has superior efficacy to paracetamol at all doses studied in this review. Novel drugs which combine paracetamol and ibuprofen formulations within the same tablet are showing encouraging results and based on limited evidence largely based on time to re‐medication, appear to be superior analgesics to the constituent drugs taken individually. The reasons for this could be related to the formulation of the combination drug, although this is by no means confirmed in the literature at present. The rationale for combined analgesia is that enhanced pain relief can potentially be achieved from two drugs with different modes of action using a lower dose and with reduced side effects, this is the basis for developing combination analgesics (Daniels 2011; Derry 2013).
Overall completeness and applicability of evidence
All of the included trials only looked at pain relief and intensity data following a single dose of the trial analgesic in a postoperative pain setting. From a clinical point of view, this model has limitations although it is the most frequently used method to assess the efficacy of analgesics. As we know, pain does continue following the initial analgesic dose and the drugs evaluated in this review are normally prescribed to be taken at a frequency of every six to eight hours (maximum of four times per day allowing for time spent sleeping). It would be of interest to know what the pain experience is following the second and subsequent doses of these medications.
All included studies used the 'third molar' pain model or 'dental pain' model to assess their outcomes. This method of pain modelling has been criticised due to not being representative of the entire population. The patients who are enrolled on to these studies would typically be:
-
aged under 30 years;
-
in good general health;
-
lacking in previous surgical interventions; and
-
physically fit and active.
These categories will not apply to the entire population (Daniels 2011).
Within the review we presented TOTPAR (total pain relief) data at two and six hours postdosing. We are aware that TOTPAR equations have only been validated for four‐ and six‐hour data (Moore 1996; Moore 1997a; Moore 1997b). Results for two hours, presented within this review, should be interpreted with caution, although from a clinical perspective it was felt that two‐hour data were more clinically relevant than four‐hour.
Quality of the evidence
As only double‐blinded (for participant and outcome assessor) randomised controlled trials were included, the risk of bias was low for two trials, unclear for two trials and high for three trials where there was concern that the type of anaesthetic given may have introduced bias and some lack of detail as to how the participants were randomised. The summary of findings tables present the overall quality of the evidence for each comparison and this was assessed as high quality for comparing paracetamol and ibuprofen; this means that further research is very unlikely to change our confidence in the estimates of the effect. The body of evidence for the proportion of patients with > 50% maximum pain relief (TOTPAR) over two and six hours, when comparing combined (ibuprofen and paracetamol) versus single drugs, was assessed as moderate quality due to these being single studies or based on high risk of bias trials; this means that further research is likely to have an important impact on our confidence in the estimate of the effect. The body of evidence for the use of rescue medication was assessed as being of high quality.
Potential biases in the review process
A thorough search was conducted to locate the included studies, it is highly unlikely that any relevant studies were missed in our search process. None of the authors are featured on any of the included studies and there are no known conflicts of interest.
Agreements and disagreements with other studies or reviews
Based on previously published Cochrane reviews using only the third molar model for assessing analgesics, 400 mg ibuprofen has a number needed to treat (NNT) of 2.3 (95% confidence interval 2.2 to 2.4) (Derry 2009; Moore 2011), and 975 mg to 1000 mg paracetamol has an NNT of 3.6 (95% confidence interval 3.2 to 4.0) (Moore 2011; Toms 2008). Therefore the conclusions from this review are in agreement with those from existing Cochrane reviews demonstrating that ibuprofen is a more effective analgesic than paracetamol at the most frequently prescribed doses. The superiority of the combined drugs over the individual drugs is also echoed by another recent Cochrane review which found that participants had a smaller chance of requiring rescue medication over eight hours if they took the combined drug compared to the individual agents (Derry 2013).
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1 Ibuprofen versus paracetamol, Outcome 1 Proportion of patients with > 50% max pain relief (TOTPAR) over 6 hours.
Comparison 1 Ibuprofen versus paracetamol, Outcome 2 Proportion of patients with > 50% max pain relief (TOTPAR) over 2 hours.
Comparison 1 Ibuprofen versus paracetamol, Outcome 3 Number of patients using rescue medication at 6 hours.
Comparison 1 Ibuprofen versus paracetamol, Outcome 4 Number of patients using rescue medication at 8 hours.
Comparison 2 Combined (ibuprofen and paracetamol) versus single drugs, Outcome 1 Proportion of patients with > 50% max pain relief (TOTPAR) over 6 hours.
Comparison 2 Combined (ibuprofen and paracetamol) versus single drugs, Outcome 2 Proportion of patients with > 50% max pain relief (TOTPAR) over 2 hours.
Comparison 2 Combined (ibuprofen and paracetamol) versus single drugs, Outcome 3 Number of patients using rescue medication at 8 hours.
| Ibuprofen versus paracetamol for pain relief following the surgical removal of lower wisdom teeth | ||||||
| Patient or population: Patients with pain after surgical removal of lower wisdom teeth Control: Paracetamol | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed1 risk | Corresponding risk | |||||
| Control | Ibuprofen | |||||
| Proportion of patients with > 50% maximum pain relief (TOTPAR) over 6 hours ‐ Ibuprofen 400 mg versus paracetamol 1000 mg | Study population | RR 1.47 | 646 | ⊕⊕⊕⊕ | When all doses considered RR = 1.45 (95% CI 1.31 to 1.61) (6 studies, 926 participants; high quality evidence) | |
| 56 per 100 | 83 per 100 | |||||
| Proportion of patients with > 50% maximum pain relief (TOTPAR) over 2 hours ‐ Ibuprofen 400 mg versus paracetamol 1000 mg | Study population | RR 1.30 | 645 | ⊕⊕⊕⊕ | When all doses considered RR = 1.29 (95% CI 1.13 to 1.46) (6 studies, 926 participants; high quality evidence) | |
| 62 per 100 | 81 per 100 | |||||
| Number of patients not using rescue medication at 6 hours (non‐event) Follow‐up: 6 hours | Study population | RR 1.50 | 542 | ⊕⊕⊕⊕ | When all doses considered RR = 1.44 (95% CI 1.13 to 1.64) (5 studies, 823 participants; high quality evidence) | |
| 50 per 100 | 75 per 100 | |||||
| Adverse events | The majority of adverse events were minor in nature and usually included nausea, vomiting, headaches and dizziness. Side effect profiles have not been included in a meta‐analysis as multiple adverse events were recorded in single patients. However, the differences in the observed adverse events for ibuprofen and paracetamol were small | |||||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 This is the median control group risk based on paracetamol being the control group. | ||||||
| Combined (ibuprofen and paracetamol) versus single drugs for pain relief after surgical removal of lower wisdom teeth | ||||||
| Patient or population: Patients with pain after surgical removal of lower wisdom teeth Control: Single drugs | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed1 risk | Corresponding risk | |||||
| Control | Combined (ibuprofen and paracetamol) | |||||
| Proportion of patients with > 50% maximum pain relief (TOTPAR) over 6 hours ‐ Paracetamol 1000 mg/ibuprofen 400 mg versus paracetamol 1000 mg or ibuprofen 400 mg | Study population | RR 1.77 | 170 | ⊕⊕⊕⊝ | ||
| 38 per 100 | 67 per 100 | |||||
| Proportion of patients with > 50% maximum pain relief (TOTPAR) over 2 hours ‐ Paracetamol 1000 mg/ibuprofen 400 mg versus paracetamol 1000 mg or ibuprofen 400 mg | Study population | RR 1.29 | 170 | ⊕⊕⊕⊝ | ||
| 37 per 100 | 48 per 100 | |||||
| Number of patients not using rescue medication at 8 hours ‐ Paracetamol 1000 mg/ibuprofen 400 mg versus paracetamol 1000 mg or ibuprofen 400 mg | Study population | RR 1.60 | 467 | ⊕⊕⊕⊕ | ||
| 50 per 100 | 80 per 100 | |||||
| Adverse events | The majority of adverse events were minor in nature and usually included nausea, vomiting, headaches and dizziness. Side effect profiles have not been included in a meta‐analysis as multiple adverse events were recorded in single patients. However, the differences in the observed adverse events for ibuprofen and paracetamol were small | |||||
| *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 This is the median control group risk based on paracetamol/ibuprofen as single drugs being the control group. 2 Quality of evidence downgraded due to single study and serious imprecision. | ||||||
|
| Daniels 2009 | Forbes 1990 | Hersch 2000* | Mehlisch 1995 | Mehlisch 2010 | Mehlisch 2010a | Olson 2001* |
| Paracetamol 500 mg
|
|
|
|
|
| √ |
|
| Paracetamol 600 mg
|
| √ |
|
|
|
|
|
| Paracetamol 1000 mg
| √ |
| √ | √ | √ | √ | √ |
| Ibuprofen 200 mg
|
|
| √ |
|
| √ |
|
| Ibuprofen 400 mg
| √ | √ | √ | √ | √ | √ | √ |
| Ibuprofen 512 mg
| √ |
|
|
|
|
|
|
| Paracetamol 250 mg/ibuprofen 100 mg
|
|
|
|
|
| √ |
|
| Paracetamol 500 mg/ibuprofen 200 mg
|
|
|
|
| √ | √ |
|
| Paracetamol 1000 mg/ibuprofen 400 mg
|
|
|
|
| √ | √ |
|
| *Liquigel formula. | |||||||
| Study | Use of rescue medication (RM) | ||||
| Mean time to RM | Use of RM (%) | Use of RM (n) | Total | Observation period | |
| Daniels 2009 | 6 hours | ||||
| Paracetamol 1000 | 44 | 35 | 80 | ||
| Ibuprofen 400 | 23 | 18 | 80 | ||
| Ibuprofen 512 | 33 | 26 | 80 | ||
| Forbes 1990 | 6 hours | ||||
| Paracetamol 600 | 3.89 hours | 81 | 29 | 36 | |
| Ibuprofen 400 | 4.63 hours | 63 | 20 | 32 | |
| Hersch 2000 | 6 hours | ||||
| Paracetamol 1000 | 51 | 32 | 63 | ||
| Ibuprofen 200 | 31 | 19 | 61 | ||
| Ibuprofen 400 | 22 | 13 | 59 | ||
| Mehlisch 1995 | 6 hours | ||||
| Paracetamol 1000 | 60 | 60 | 101 | ||
| Ibuprofen 400 | 26 | 25 | 98 | ||
| Mehlisch 2010 | 8 hours | ||||
| Paracetamol 1000 | 261 minutes | 71 | 24 | 34 | |
| Ibuprofen 400 | 296 minutes | 68 | 47 | 69 | |
| Paracetamol 1000/ibuprofen 400 | 376 minutes | 31 | 21 | 67 | |
| Paracetamol 500/ibuprofen 200 | 329 minutes | 61 | 20 | 33 | |
| Mehlisch 2010a | 8 hours in phase 1 | ||||
| Paracetamol 500 | 74 | 56 | 76 | ||
| Paracetamol 1000 | 69 | 51 | 74 | ||
| Ibuprofen 200 | 39 | 29 | 75 | ||
| Ibuprofen 400 | 28 | 21 | 74 | ||
| Paracetamol 250/ibuprofen 100 | 38 | 27 | 71 | ||
| Paracetamol 1000/ibuprofen 400 | 28 | 42 | 149 | ||
| Paracetamol 500/ibuprofen 200 | 22 | 31 | 143 | ||
| Olson 2001 | 6 hours | ||||
| Paracetamol 1000 | 38 | 25 | 66 | ||
| Ibuprofen 400 | 21 | 14 | 67 | ||
| Study | Adverse events (AE) | |||
| Total AE (n) | Total AE % | Serious/Severe AE | Serious/Severe AE % | |
| Daniels 2009 | ||||
| Paracetamol 1000 | 25 (81) | 31 | 3 | 12 |
| Ibuprofen 400 | 19 (80) | 24 | 1 | 3 |
| Ibuprofen 512 | 24 (80) | 30 | 2 | 6 |
| Forbes 1990 | ||||
| Paracetamol 600 | 5 (41) | 12 | Not specified | Not specified |
| Ibuprofen 400 | 8 (43) | 19 | Not specified | Not specified |
| Hersch 2000 | ||||
| Paracetamol 1000 | 12 (63) | 19 | 0 | 0 |
| Ibuprofen 200 | 7 (61) | 11 | 0 | 0 |
| Ibuprofen 400 | 4 (59) | 7 | 0 | 0 |
| Mehlisch 1995 | ||||
| Paracetamol 1000 | 17 (101) | 17 | 0 | 0 |
| Ibuprofen 400 | 12 (98) | 12 | 0 | 0 |
| Mehlisch 2010 | ||||
| Paracetamol 1000 | 24 (34) | 71 | 11 | 46 |
| Ibuprofen 400 | 39 (69) | 57 | 14 | 36 |
| Paracetamol 1000/ibuprofen 400 | 38 (67) | 57 | 11 | 29 |
| Paracetamol 500/ibuprofen 200 | 14 (33) | 42 | 6 | 43 |
| Mehlisch 2010a | ||||
| Paracetamol 500 | 38 (76) | 50 | 7 | 18 |
| Paracetamol 1000 | 30 (74) | 40 | 5 | 17 |
| Ibuprofen 200 | 31 (75) | 41 | 7 | 23 |
| Ibuprofen 400 | 33 (74) | 45 | 5 | 15 |
| Paracetamol 250/ibuprofen 100 | 22 (71) | 31 | 8 | 36 |
| Paracetamol 1000/ibuprofen 400 | 44 (149) | 30 | 15 | 34 |
| Paracetamol 500/ibuprofen 200 | 51 (143) | 36 | 7 | 14 |
| Olson 2001 | ||||
| Paracetamol 1000 | 10 (66) | 15 | 1 | 10 |
| Ibuprofen 400 | 7 (67) | 11 | 2 | 29 |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Proportion of patients with > 50% max pain relief (TOTPAR) over 6 hours Show forest plot | 6 | 926 | Risk Ratio (M‐H, Random, 95% CI) | 1.45 [1.31, 1.61] |
| 1.1 Ibuprofen 200 mg versus paracetamol 1000 mg | 1 | 92 | Risk Ratio (M‐H, Random, 95% CI) | 1.29 [0.90, 1.84] |
| 1.2 Ibuprofen 400 mg versus paracetamol 1000 mg | 5 | 646 | Risk Ratio (M‐H, Random, 95% CI) | 1.47 [1.28, 1.69] |
| 1.3 Ibuprofen 400 mg versus paracetamol 600 mg | 1 | 68 | Risk Ratio (M‐H, Random, 95% CI) | 2.41 [1.13, 5.16] |
| 1.4 Ibuprofen 512 mg versus paracetamol 1000 mg | 1 | 120 | Risk Ratio (M‐H, Random, 95% CI) | 1.43 [1.15, 1.78] |
| 2 Proportion of patients with > 50% max pain relief (TOTPAR) over 2 hours Show forest plot | 6 | 926 | Risk Ratio (M‐H, Random, 95% CI) | 1.29 [1.13, 1.46] |
| 2.1 Ibuprofen 512 mg versus paracetamol 1000 mg | 1 | 120 | Risk Ratio (M‐H, Random, 95% CI) | 1.28 [0.98, 1.67] |
| 2.2 Ibuprofen 400 mg versus paracetamol 1000 mg | 5 | 645 | Risk Ratio (M‐H, Random, 95% CI) | 1.30 [1.09, 1.55] |
| 2.3 Ibuprofen 200 mg versus paracetamol 1000 mg | 1 | 93 | Risk Ratio (M‐H, Random, 95% CI) | 1.09 [0.85, 1.41] |
| 2.4 Ibuprofen 400 mg versus paracetamol 600 mg | 1 | 68 | Risk Ratio (M‐H, Random, 95% CI) | 1.74 [0.96, 3.14] |
| 3 Number of patients using rescue medication at 6 hours Show forest plot | 5 | 823 | Risk Ratio (M‐H, Random, 95% CI) | 1.44 [1.26, 1.64] |
| 3.1 Ibuprofen 200 mg versus paracetamol 1000 mg | 1 | 93 | Risk Ratio (M‐H, Random, 95% CI) | 1.38 [0.94, 2.02] |
| 3.2 Ibuprofen 400 mg versus paracetamol 1000 mg | 4 | 542 | Risk Ratio (M‐H, Random, 95% CI) | 1.50 [1.25, 1.79] |
| 3.3 Ibuprofen 512 mg versus paracetamol 1000 mg | 1 | 120 | Risk Ratio (M‐H, Random, 95% CI) | 1.17 [0.86, 1.60] |
| 3.4 Ibuprofen 400 mg versus paracetamol 600 mg | 1 | 68 | Risk Ratio (M‐H, Random, 95% CI) | 1.93 [0.87, 4.30] |
| 4 Number of patients using rescue medication at 8 hours Show forest plot | 2 | 402 | Risk Ratio (M‐H, Fixed, 95% CI) | 2.02 [1.57, 2.60] |
| 4.1 Ibuprofen 200 mg versus paracetamol 500 mg | 1 | 75 | Risk Ratio (M‐H, Fixed, 95% CI) | 2.36 [1.31, 4.25] |
| 4.2 Ibuprofen 400 mg versus paracetamol 500 mg | 1 | 75 | Risk Ratio (M‐H, Fixed, 95% CI) | 2.77 [1.57, 4.89] |
| 4.3 Ibuprofen 200 mg versus paracetamol 1000 mg | 1 | 75 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.87 [1.10, 3.17] |
| 4.4 Ibuprofen 400 mg versus paracetamol 1000 mg | 2 | 177 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.66 [1.11, 2.48] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Proportion of patients with > 50% max pain relief (TOTPAR) over 6 hours Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 1.1 Paracetamol 1000 mg/ibuprofen 400 mg versus paracetamol 1000 mg or ibuprofen 400 mg | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 2 Proportion of patients with > 50% max pain relief (TOTPAR) over 2 hours Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 2.1 Paracetamol 1000 mg/ibuprofen 400 mg versus paracetamol 1000 mg or ibuprofen 400 mg | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 3 Number of patients using rescue medication at 8 hours Show forest plot | 2 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 3.1 Paracetamol 1000 mg/ibuprofen 400 mg versus paracetamol 1000 mg or ibuprofen 400 mg | 2 | 467 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.60 [1.36, 1.88] |
