Oral analgesia for relieving post-caesarean pain.

Cochrane Database of Systematic Reviews

Oral analgesia for relieving post-caesarean pain (Review) Mkontwana N, Novikova N

Mkontwana N, Novikova N. Oral analgesia for relieving post-caesarean pain. Cochrane Database of Systematic Reviews 2015, Issue 3. Art. No.: CD010450. DOI: 10.1002/14651858.CD010450.pub2.

www.cochranelibrary.com

Oral analgesia for relieving post-caesarean pain (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

TABLE OF CONTENTS HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.1. Comparison 1 Opioid versus placebo, Outcome 1 Need for additional pain relief with a different drug. Analysis 1.2. Comparison 1 Opioid versus placebo, Outcome 2 Adverse drug effects. . . . . . . . . . . . Analysis 2.1. Comparison 2 Non-opioid analgesics versus placebo, Outcome 1 Need for additional pain relief. . . . Analysis 2.2. Comparison 2 Non-opioid analgesics versus placebo, Outcome 2 Maternal drug effects. . . . . . . Analysis 3.1. Comparison 3 Combination versus placebo, Outcome 1 Need for additional pain relief. . . . . . . Analysis 3.2. Comparison 3 Combination versus placebo, Outcome 2 Maternal drug effects. . . . . . . . . . Analysis 4.1. Comparison 4 Opioid versus non-opioid, Outcome 1 Need for additional pain relief with a different drug. Analysis 4.2. Comparison 4 Opioid versus non-opioid, Outcome 2 Maternal adverse effects. . . . . . . . . . Analysis 5.1. Comparison 5 Opioid analgesics versus combination analgesics, Outcome 1 Need for additional pain relief. Analysis 5.2. Comparison 5 Opioid analgesics versus combination analgesics, Outcome 2 Maternal adverse effects. . Analysis 6.1. Comparison 6 Non-opioid versus combination analgesics, Outcome 1 Need for additional pain relief with a different drug. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 7.1. Comparison 7 Non-opioid analgesics versus placebo (subgroup analysis by high and low doses of the same drug), Outcome 1 Need for additional pain relief. . . . . . . . . . . . . . . . . . . . . . Analysis 8.1. Comparison 8 Opioid analgesics versus placebo (subgroup analysis by high and low doses of the same drug), Outcome 1 Need for additional pain relief. . . . . . . . . . . . . . . . . . . . . . . . . WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . . INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


1 1 3 3 5 5 9 10 12 13 16 18 18 18 23 40 42 43 44 45 46 47 47 48 49 49 50 51 52 52 52 53 53 53

i

[Intervention Review]

Oral analgesia for relieving post-caesarean pain Nondumiso Mkontwana1 , Natalia Novikova1 1

Department of Obstetrics and Gynaecology, East London Hospital Complex, Walter Sisulu University, East London, South Africa

Contact address: Nondumiso Mkontwana, Department of Obstetrics and Gynaecology, East London Hospital Complex, Walter Sisulu University, East London, Eastern Cape, 5200, South Africa. [email protected]. Editorial group: Cochrane Pregnancy and Childbirth Group. Publication status and date: New, published in Issue 3, 2015. Review content assessed as up-to-date: 31 July 2014. Citation: Mkontwana N, Novikova N. Oral analgesia for relieving post-caesarean pain. Cochrane Database of Systematic Reviews 2015, Issue 3. Art. No.: CD010450. DOI: 10.1002/14651858.CD010450.pub2. Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

ABSTRACT Background Oral analgesia is a convenient and widely used form of pain relief following caesarean section. It includes various medications used at different doses alone or in adjunction to other form of analgesia. Objectives To determine the effectiveness, safety and cost-effectiveness of oral analgesia for post-caesarean pain relief. Search methods We searched the Cochrane Pregnancy and Childbirth Group’s Trials Register (31 July 2014) and reference lists of retrieved studies. Selection criteria Randomised controlled trials (RCTs). Cluster-randomised trials were eligible for inclusion but none were identified. Quasi-randomised and cross-over trials were not eligible for inclusion. Interventions included oral medication given to women for post-caesarean pain relief compared with oral medication, or placebo/no treatment. Data collection and analysis Two review authors independently assessed for inclusion all the potential studies and independently assessed trial quality, extracted the data using the agreed data extraction form, and checked them for accuracy. Main results Eight small trials involving 962 women (out of 13 included trials) contributed data to the analysis, of which only four trials had low risk of bias. None of the included studies reported on ’adequate pain relief ’, which is one of this review’s primary outcomes. 1. Opiod analgesics versus placebo Based on one trial involving 120 women, the effect of opioids versus placebo was not significant in relation to the need for additional pain relief (primary outcome) (risk ratio (RR) 0.33, 95% confidence interval (CI) 0.06 to 1.92), and the effect in terms of adverse drug effects outcomes was also uncertain (RR 6.58, 95% CI 0.38 to 113.96). Oral analgesia for relieving post-caesarean pain (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

1

Low (75 mg) and high (150 mg) doses of tramadol had a similar effect on the need for additional pain relief (RR 0.67, 95% CI 0.12 to 3.78 and RR 0.14, 95% CI 0.01 to 2.68, respectively, one study, 80 women). 2. Non-opioid analgesia versus placebo The confidence interval for the lower requirement for additional analgesia (primary outcome) with the non-opioid analgesia group was wide and includes little or no effect (average RR 0.70, 95% CI 0.48 to 1.01, six studies, 584 women). However, we observed substantial heterogeneity due to the variety of non-opioid drugs used (I2 = 85%). In a subgroup analysis of different drugs, only gabapentin use resulted in less need for additional pain relief (RR 0.34, 95% CI 0.23 to 0.51, one trial, 126 women). There was no difference in need for additional pain relief with the use of celexocib, ibuprofen, ketoprofen, naproxen, paracetamol. Maternal drug effects were more common with the use of non-opioid analgesics (RR 11.12, 95% CI 2.13 to 58.22, two trials, 267 women). Gabapentin 300 mg (RR 0.25, 95% CI 0.13 to 0.49, one study, 63 women) and 600 mg (RR 0.44, 95% CI 0.27 to 0.71, one study, 63 women) as well as ketoprofen 100 mg (RR 0.55, 95% CI 0.39 to 0.79, one study 72 women) were both more effective than placebo with respect to the need for additional pain relief. However, the 50 mg ketoprofen group and the placebo group did not differ in terms of the number of women requiring additional pain relief (RR 0.82, 95% CI 0.64 to 1.07, one study, 72 women). 3. Combination analgesics versus placebo Our pooled analysis for the effect of combination analgesics on the need for additional pain relief was RR 0.70 (95% CI 0.35 to 1.40, three trials, 242 women, I2 = 69%). When comparing different drugs within the combination oral analgesics versus placebo comparison we observed subgroup differences (P = 0.05; I² = 65.8%). One trial comparing paracetamol plus codeine versus placebo resulted in fewer women requiring additional pain relief (RR 0.44, 95% CI 0.23 to 0.82, one trial, 65 women). However, there were no differences in the the number of women requiring additional pain relief when comparing paracetamol plus oxycodone versus placebo, or paracetamol plus propoxyphene (RR 1.00, 95% CI 0.78 to 1.28, one trial, 96 women and RR 0.65, 95% CI 0.11 to 3.69, one trial, 81 women, respectively). Maternal drug effects were more common in combination analgesics group versus placebo (RR 13.18, 95% CI 2.86 to 60.68, three trials, 252 women). 4. Opioid analgesics versus non-opioid analgesics The confidence interval for the effect on additional pain relief between opioid and non-opioid drugs was very wide (RR 0.51, 95% CI 0.07 to 3.51, one trial, 121 women). Side effects were more common with the use opioids versus non-opioids analgesics (RR 2.32, 95% CI 1.15 to 4.69, two trials 241 women). 5. Opioid analgesics versus combination analgesics There was no difference in need for additional pain relief in opioid analgesics versus combination analgesics based on one study involving 121 women comparing tramadol and paracetamol plus propoxyphene (RR 0.51, 95% CI 0.07 to 3.51). Maternal adverse effects also did not differ between the two groups (RR 6.74, 95% CI 0.39 to 116.79). 6. Non-opioid versus combination analgesics The need for additional pain relief was greater in the group of women who received non-opoid analgesics (RR 0.87, 95% CI 0.81 to 0.93, one trial, 192 women) compared with the group of women who received combination analgesics. Secondary outcomes not reported in the included studies No data were found on the following secondary outcomes: number of days in hospital post-operatively, re-hospitalisation due to incisional pain, fully breastfeeding on discharge, mixed feeding at discharge, incisional pain at six weeks after caesarean section, maternal post partum depression, effect (negative) on mother and baby interaction and cost of treatment. Authors’ conclusions Eight trials with 962 women were included in the analysis, but only four trials were of high quality. All the trials were small. We carried out subgroup analysis for different drugs within the same group and for high versus low doses of the same drug. However, the relatively few studies (one to two trials) and numbers of women (40 to 136) limits the reliability of these subgroup analyses. Due to limited data available no conclusions can be made regarding the safest and the most effective form of oral analgesia for postcaesarean pain. Further studies are necessary. Oral analgesia for relieving post-caesarean pain (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

2

PLAIN LANGUAGE SUMMARY Pain tablets taken by mouth for post-caesarean pain Tablets are convenient and easy to take to ease the pain after caesarean section, which involves cutting through the abdomen and uterus to deliver the baby. We aimed to assess the effectiveness, safety and cost-effectiveness of different types of tablets for the pain. Different types of pain tablets relieve the pain in different ways. Opioids decrease the feeling of pain, decrease reaction to pain as well as increase pain tolerance by their action on the nervous system. Some non-opioid pain tablets act on the tissues to reduce the response to the inflammatory substances released at the site of tissue damage. Combination drugs (such as paracetamol and codeine) may have more pronounced effects because of the different mechanism of action of their components. We do not know how some other tablets, such as alpha-2 agonists (clonidine) and gabapentin (usually used for nerve pain that follows shingles and long-term pain) relieve pain. Good pain control may shorten the time spent in hospital after caesarean section, improve satisfaction and reduce healthcare costs. Searching the literature we found 63 articles assessing post-caesarean pain but only 13 studies met our inclusion criteria, of which only eight studies (involving 962 women) reported on outcomes that we assessed in this review. Of the eight trials contributing data to our review, only four trials had low risk of bias. All of the included studies involved small numbers of women (40 to 136 women). We compared non-opioid, opioid and combination oral tablets with placebo or no treatment. There was insufficient evidence to determine the effect of additional pain killers (tablets or injections) among women who took opioid or non-opioid or combination pain killers in comparison to placebo. There was also insufficient evidence to establish the effect of opioid versus non-opioid drugs and opioid versus combination drugs on requirement for additional pain relief. When assessing different non-opioid drugs individually, we found some evidence that gabapentin resulted in less need for additional pain relief in comparison to placebo, but the analysis of data for other tablets (celexocib, ibuprofen, ketoprofen, naproxen, paracetamol) gave more uncertain results. The use of paracetamol plus codeine resulted in less need for additional pain relief in comparison to placebo. We found that high and low dose of gabapentin, and high dose of ketoprofen were more effective than placebo in relation to the need for additional pain relief. On the other hand, low dose of ketoprofen as well as high and low dose of tramadol did not differ with placebo with in effect on the need for additional pain relief. However, it is important to note that these additional analyses (i.e. different drugs and different doses) are based on relatively few studies (one to two trials) and numbers of women (40 to 136). Women developed more side effects including nausea, vomiting and drowsiness with the use of opioid, non-opioid or combination painkillers in comparison to placebo or no treatment. All trials used additional pain relief as a standard and for breakthrough pain and they all used different drugs. No studies reported on the following outcomes: adequate pain relief, number of days in hospital post operatively, re-hospitalisation due to incisional pain, fully breast feeding on discharge, mixed feeding at discharge, maternal postpartum depression and cost-effectiveness of the studied interventions. Due to limited data available no conclusions can be made regarding the safest and the most effective form of oral analgesia for postcaesarean pain. Further studies are necessary. More research is needed to compare different types of pain killers and in relation to different outcomes such as safety, efficacy and cost.

BACKGROUND Caesarean section is a common surgical procedure performed to deliver babies through an incision of the abdomen and uterus when vaginal delivery is contraindicated for maternal or fetal reasons or, in some cases upon request of the pregnant mother. It can either be done as an emergency or electively. The rate of caesarean section has been increasing around the world (Delbaere 2012). The rate of caesarean section differs around the world between 2% in least

developed countries and 21% in developed countries and reaching 45.9% in Brazil (Betran 2007; Getahun 2009; Gibbons 2010). In the USA, caesarean section rates increased, from 20.7% in 1996 to 31.1% in 2006 for women of all ages, race, gestational ages, and in all states (MacDorman 2008). Caesarean section is associated with more postpartum pain than vaginal birth (Kainu 2010) and leads to more acute and chronic postpartum pain (Ingrid 2006). Postpartum pain has various neg-


3

ative consequences including: • it causes significant discomfort to women, which can lead to difficulties in mobility and subsequent problems, such as, an increased risk of venous thrombosis, and by interfering with optimal interaction with the newborn in the immediate postpartum period; • shallow breathing and splinting may result in atelectasis and predispose to pneumonia; • pain has psychological sequelae (Berghella 2012; DiMatteo 1996); • it may reduce the ability of the mother to initiate or continue with breastfeeding effectively (Gadsden 2005). Effective post-caesarean pain relief is important to avoid the abovementioned problems. Different interventions have been proposed for post-caesarean pain relief, e.g. oral, intravenous and rectal analgesia with various drugs and various doses, regional analgesia, transversus abdominis plane block, wound infiltration and combinations of the above-mentioned interventions (Ismail 2012). The review aimed to compare the effectiveness and safety of different classes of oral analgesia for post-caesarean pain relief.

1. Opioid analgesics • Natural opioids (codeine, dihydrocodeine, morphine) • Diphenylpropylamine derivatives (dextro propoxyphene, dipipanone) • Other opioids (tramadol, tilidine) 2. Non-opioid analgesics A. Para-aminophenol derivatives (paracetamol or acetaminophen) B. Non-selective non-steroidal anti-inflammatory drugs (NSAIDs) • Acetylated salicylates (aspirin or acetylsalicylic acid) • Propionic acid (ibuprofen, ketoprofen) • Acetic acids (diclofenac) • Oxicams (meloxicam) • Fenamates (mefenamic acid) C. Alpha-2 agonists (clonidine) D. Anti-convulsants (gabapentin)

3. Combination drugs (e.g. paracetamol/codeine, paracetamol/tramadol, paracetamol/codeine/ibuprofen, etc.) Oral analgesics have various side effects. The adverse effects of these drugs include nausea, vomiting, constipation, diarrhoea, drowsiness, respiratory depression, pruritis (itch), rash, fluid retention.

Description of the condition Post-caesarean pain is a result of incisional pain as well as pain from the uterus (e.g. uterine contraction after birth). Incisional pain is experienced by women following caesarean section due to tissue trauma from surgical incision, dissection and burns or direct nerve damage from nerve transection, stretching or compression (Kelly 2001). Tissue trauma causes release of local inflammatory mediators that can produce augmented sensitivity to stimuli in the area surrounding an injury, i.e. hyperalgesia or misperception pain to non-noxious stimuli. The patient senses pain through the afferent pathway. Pharmacologic agents target these pathways ( Woolf 1993). Pain varies in intensity and the onset of the post-operative pain depends on form of anaesthesia used during the procedure. This pain almost always requires some form of analgesia (Kodali 2012).

Description of the intervention Oral analgesia is a simple, easy to administer, well-tolerated and cost-effective type of pain relief that is offered to women after caesarean section. Oral analgesia for post-caesarean pain includes the following ( SAMF 2012; Solomon 2012).

How the intervention might work Different types of oral analgesics have different ways of achieving an analgesic (painkiller) effect. Analgesic effects of opioids are due to decreased perception of pain, decreased reaction to pain as well as increased pain tolerance. Opioids act through binding to specific opioid receptors in the nervous system and other tissues. Opioids reduce the perception of pain by activating pain-inhibitory neurons and inhibiting pain transmission neurons (Chahl 1996). Non-opioid analgesics act by reducing the nocioreceptive response to the endogenous inflammatory mediators released at the sites of tissue damage (Kuo 2006). NSAIDs block cyclo-oxygenase (COX), an enzyme responsible for the synthesis of prostaglandins (Chandrasekharan 2002). Pharmacological inhibition of COX can provide relief from symptoms of inflammation and pain by inhibition of prostaglandin synthesis. Combination drugs (e.g. paracetamol/codeine, paracetamol/ ibuprofen) exhibit enhanced effects due to the different mechanism of action of their components (SAMF 2012). The exact mechanism of analgesic effect of alpha-2 agonists is unknown, though the release of acetylcholine may play a role (Gordh 1989; Kodali 2012). There are reports suggesting that alpha-2 agonists such as clonidine and dexmedetomidine have a


4

potent analgesic response, and that their potency is increased by concomitant opioid therapy (Kodali 2012). Alpha-2 agonists have been reported to decrease post-operative pain (Hidalgo 2005; Sung 2000). Anticonvulsant agents such as gabapentin are gamma-aminobutyric acid analogues and have also been reported to have analgesic effects as well as opioid-sparing effects (Kodali 2012). Although the main use of anticonvulsants is for chronic pain, there are reports supporting their adjunctive role in post-operative pain (Mathiesen 2007; Moore 2009).

Why it is important to do this review Many drugs with various mechanisms of action are used for postcaesarean pain relief. Although the response to pain relief is sometimes believed to be individual, it is very important to establish the most effective with the least adverse effects type of oral analgesia for women after caesarean section. Optimal pain control post-caesarean section will benefit not only the mother and her baby, but also a healthcare system. Optimal pain control may shorten the time spent in hospital after caesarean section and, therefore, reduce healthcare costs (Shang 2003).

Types of participants All women requiring pain relief in the early postpartum period following caesarean section.

Types of interventions Interventions included oral medication given to women for postcaesarean pain relief. The comparisons are: 1. opioid analgesics versus placebo/no drug treatment; 2. non-opioid analgesics versus placebo/no drug treatment; 3. combination drugs verus placebo/no drug treatment; 4. opioid analgesics versus non-opioid analgesics; 5. opioid analgesics versus combination analgesics; 6. non-opioid analgesics versus combination analgesics.

Types of outcome measures We assessed outcomes developed in the Cochrane generic protocol on drugs for perineal pain in the early postpartum period (Chou 2009).

Primary outcomes

OBJECTIVES To determine the effectiveness (adequate pain relief as reported by the woman, or by determination of more than 50% relief of pain (either stated by the woman or calculated using a formula), or the need for additional pain relief with a different drug, re-admission due to incisional pain, incisional pain at six weeks after surgery), safety (e.g. adverse effects of drugs, effect on breastfeeding, depression, mother and child interaction) and cost-effectiveness of oral analgesics for relieving post-caesarean pain.

METHODS

Criteria for considering studies for this review

1. Adequate pain relief as reported by the woman, or by determination of more than 50% relief of pain (as either stated by the woman or calculated using a formula)*. 2. Need for additional pain relief with a different drug.

Secondary outcomes

1. Maternal drug adverse effects, e.g. nausea, vomiting, sedation, constipation, diarrhoea, drowsy, sleepy, psychological impact. 2. Number of days in the hospital post-operatively. 3. Rehospitalisation due to incisional pain. 4. Fully breastfeeding at discharge. 5. Mixed feeding at discharge. 6. Incisional pain at six weeks after caesarean section. 7. Maternal postpartum depression. 8. Effect (negative) on mother and baby interaction. 9. Cost of treatment. * Assessment of 50% pain relief via SPID (summed pain intensity difference) scores (1.23 x SPID %max - 2.3 = proportion with 50%) (Cooper 1991; Moore 1967; Moore 1967a).

Types of studies We included randomised controlled trials (RCTs). Cluster-randomised trials were eligible for inclusion but none were identified. Quasi-randomised and cross-over trials were not eligible for inclusion.

Search methods for identification of studies The following methods section of this review is based on a standard template used by the Cochrane Pregnancy and Childbirth Group.


5

Electronic searches We contacted the Trials Search Co-ordinator to search the Cochrane Pregnancy and Childbirth Group’s Trials Register (31 July 2014). The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co-ordinator and contains trials identified from: 1. monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL); 2. weekly searches of MEDLINE (Ovid); 3. weekly searches of Embase (Ovid); 4. handsearches of 30 journals and the proceedings of major conferences; 5. weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts. Details of the search strategies for CENTRAL, MEDLINE and Embase, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group. Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co-ordinator searches the register for each review using the topic list rather than keywords.

Searching other resources We searched reference lists of retrieved studies. We did not apply any language or date restrictions.

Data collection and analysis The following methods section of this review is based on a standard template used by the Cochrane Pregnancy and Childbirth Group.

Selection of studies Two review authors independently assessed for inclusion all the potential studies we identify as a result of the search strategy. We resolved any disagreement through discussion or, if required, we consulted a third person. Any studies published in an abstract form were included if they satisfied the inclusion criteria. We contacted the authors of such studies for additional information if necessary.

Data extraction and management We designed a form to extract data. For eligible studies, two review authors extracted the data using the agreed form. We resolved discrepancies through discussion (had it been necessary, we would

have consulted a third person if we were unable to resolve discrepancies through discussion). We entered data into Review Manager software (RevMan 2014) and checked for accuracy. When information regarding any of the above is unclear, we attempted to contact authors of the original reports to provide further details. Assessment of risk of bias in included studies Two review authors independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreement by discussion or by involving a third assessor. (1) Random sequence generation (checking for possible selection bias)

We described for each included study the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups. We assessed the method as: • low risk of bias (any truly random process, e.g. random number table; computer random number generator); • high risk of bias (any non-random process, e.g. odd or even date of birth; hospital or clinic record number); • unclear risk of bias. (2) Allocation concealment (checking for possible selection bias)

We described for each included study the method used to conceal allocation to interventions prior to assignment and assessed whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment. We assessed the methods as: • low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes); • high risk of bias (open random allocation; unsealed or nonopaque envelopes, alternation; date of birth); • unclear risk of bias. (3.1) Blinding of participants and personnel (checking for possible performance bias)

We described for each included study the methods used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. We considered that studies were at low risk of bias if they were blinded, or if we judged that the lack of blinding would be unlikely to affect results. We assessed blinding separately for different outcomes or classes of outcomes. We assessed the methods as: • low, high or unclear risk of bias for participants; • low, high or unclear risk of bias for personnel.


6

(3.2) Blinding of outcome assessment (checking for possible detection bias)

(6) Other bias (checking for bias due to problems not covered by (1) to (5) above)

We described for each included study the methods used, if any, to blind outcome assessors from knowledge of which intervention a participant received. We assessed blinding separately for different outcomes or classes of outcomes. We assessed methods used to blind outcome assessment as: • low, high or unclear risk of bias.

We described for each included study any important concerns we have about other possible sources of bias. We assessed whether each study was free of other problems that could put it at risk of bias: • low risk of other bias; • high risk of other bias; • unclear whether there is risk of other bias.

(4) Incomplete outcome data (checking for possible attrition bias due to the amount, nature and handling of incomplete outcome data)

We described for each included study, and for each outcome or class of outcomes, the completeness of data including attrition and exclusions from the analysis. We stated whether attrition and exclusions were reported and the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported, or could be supplied by the trial authors, we re-included missing data in the analyses which we undertook. We excluded the data from the analysis if more than 20% of data were missing for primary or secondary outcomes. We assessed methods as: • low risk of bias (e.g. no missing outcome data; missing outcome data balanced across groups); • high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; ‘as treated’ analysis done with substantial departure of intervention received from that assigned at randomisation); • unclear risk of bias.

(7) Overall risk of bias

We made explicit judgements about whether studies were at high risk of bias, according to the criteria given in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). With reference to (1) to (6) above, we assessed the likely magnitude and direction of the bias and whether we considered it was likely to* impact on the findings. We explored the impact of the level of bias through undertaking sensitivity analyses - see Sensitivity analysis.

Measures of treatment effect

Dichotomous data

For dichotomous data, we present results as summary risk ratio with 95% confidence intervals.

Continuous data

For continuous data, we planned to use the mean difference if outcomes were measured in the same way between trials. In future updates of this review we will use the standardised mean difference to combine trials that measure the same outcome, but use different methods.

Unit of analysis issues (5) Selective reporting (checking for reporting bias)

We described for each included study how we investigated the possibility of selective outcome reporting bias and what we found. We assessed the methods as: • low risk of bias (where it is clear that all of the study’s prespecified outcomes and all expected outcomes of interest to the review have been reported); • high risk of bias (where not all the study’s pre-specified outcomes have been reported; one or more reported primary outcomes were not pre-specified; outcomes of interest are reported incompletely and so cannot be used; study fails to include results of a key outcome that would have been expected to have been reported); • unclear risk of bias.

Cluster-randomised trials

We did not identify any cluster-randomised trials for inclusion. In future updates of this review, if we identify any cluster-randomised trials we will include cluster-randomised trials in the analyses along with individually-randomised trials. We will adjust their standard errors using the methods described in the Cochrane Handbook for Systematic Reviews of Interventions [Section 16.3.4 or 16.3.6] (Higgins 2011) using an estimate of the intra cluster correlation co-efficient (ICC) derived from the trial (if possible), from a similar trial or from a study of a similar population. If we use ICCs from other sources, we will report this and conduct sensitivity analyses to investigate the effect of variation in the ICC. If we identify both


7

cluster-randomised trials and individually-randomised trials, we plan to synthesise the relevant information. We will consider it reasonable to combine the results from both if there is little heterogeneity between the study designs and the interaction between the effect of intervention and the choice of randomisation unit is considered to be unlikely. We will also acknowledge heterogeneity in the randomisation unit and perform a subgroup analysis to investigate the effects of the randomisation unit.

Cross-over trials

We did not find any cross-over studies. We do not anticipate any cross-over studies on this topic. However, should we find any, we will exclude them.

Trials with multiple treatment groups

Studies with multi-group analysis were included in the review. Only the groups relevant to this review were included into the analysis. We combined all relevant experimental intervention groups of the study into a single group and all relevant control intervention groups into a single control group. For dichotomous outcomes, both the sample sizes and the numbers of people with events were summed across groups. For continuous outcomes, we planned to combine means and standard deviations using methods described in Chapter 7 (Section 7.7.3.8) of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). If considered necessary, we included each pair-wise comparison separately, but with shared intervention groups divided out approximately evenly among the comparisons (Higgins 2011). In the studies with two interventions groups and one control group we divided the number of participants and effects by half for dichotomous outcome. If continuous outcomes had been measured, we would have left the mean difference and standard deviation the same as described in Chapter 16 (Section 16.5.4) of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

Dealing with missing data For included studies, we noted levels of attrition. We explored the impact of including studies with high levels of missing data in the overall assessment of treatment effect by using sensitivity analysis. For all outcomes, we carried out analyses, as far as possible, on an intention-to-treat basis, i.e. we attempted to include all participants randomised to each group in the analyses, and all participants were analysed in the group to which they were allocated, regardless of whether or not they received the allocated intervention. If more than 20% of data were missing for primary or secondary outcomes, we excluded the results of that study from analysis.

Assessment of heterogeneity We assessed statistical heterogeneity in each meta-analysis using the Tau², I² and Chi² statistics. We regarded heterogeneity as substantial if the I² was greater than 30% and either the Tau² was greater than zero, or there was a low P value (less than 0.10) in the Chi² test for heterogeneity.

Assessment of reporting biases Where we suspected reporting bias (see ’Selective reporting bias’ above), we attempted to contact study authors to ask them to provide missing outcome data. Where this was not possible, and the missing data were thought to introduce serious bias, the impact of including such studies in the overall assessment of results was explored by a sensitivity analysis. In future updates of this review, if there are 10 or more studies in the a meta-analysis, we will investigate reporting biases (such as publication bias) using funnel plots. We will assess funnel plot asymmetry visually. If asymmetry is suggested by a visual assessment, we will perform exploratory analyses to investigate it.

Data synthesis We carried out statistical analysis using the Review Manager software (RevMan 2014). We used fixed-effect meta-analysis for combining data where it was reasonable to assume that studies were estimating the same underlying treatment effect: i.e. where trials were examining the same intervention, and the trials’ populations and methods were judged sufficiently similar. If there was clinical heterogeneity sufficient to expect that the underlying treatment effects differed between trials, or if substantial statistical heterogeneity was detected, we used random-effects meta-analysis to produce an overall summary, if an average treatment effect across trials was considered clinically meaningful. The random-effects summary was treated as the average range of possible treatment effects and we discussed the clinical implications of treatment effects differing between trials. If the average treatment effect was not clinically meaningful, we did not combine trials. Where we used random-effects analyses, the results are presented as the average treatment effect with 95% confidence intervals, and the estimates of Tau² and I².

Subgroup analysis and investigation of heterogeneity For outcomes with substantial heterogeneity, we investigated it using subgroup analyses and sensitivity analyses. We considered whether an overall summary was meaningful, and if it was, used random-effects analysis to produce it. We planned the following subgroup analyses. 1. Drugs compatible with breastfeeding versus those that are not compatible with breastfeeding because they have adverse effects on the infant.


8

2. Primiparous versus multiparous women. 3. Emergency versus elective caesarean section. 4. Primary versus repeat caesarean section. 5. Different drugs within the same group. 6. High and low doses of the same drug. Only two subgroup analyses were possible, e.g. different drugs within the same group and high and low doses of the same drug. No other planned subgroup analysis were possible due to insufficient data. All trials used baseline analgesia. Our other planned subgroup analyses will be performed the in future updates of this review. Subgroup analyses were restricted to primary outcomes of the review. We assessed subgroup differences by interaction tests available within RevMan (RevMan 2014). We reported the results of subgroup analyses quoting the Chi² statistic and P value, and the interaction test I² value.

Sensitivity analysis Planned sensitivity analyses were not performed due to insufficient data. In future updates, we will carry out sensitivity analysis to explore the effect of trial quality for primary outcomes in the

review. If the included trials have a high risk or unclear risk of selection bias or attrition bias we will explore this by conducting sensitivity analysis. We will also carry out sensitivity analysis to explore the effects of fixed-effect or random-effects analyses for outcomes with statistical heterogeneity.

RESULTS

Description of studies

Results of the search The search of the Cochrane Pregnancy and Childbirth Group’s Trials Register retrieved 63 reports (see: Figure 1). After assessing the full text and grouping together reports of the same trial, there were 54 studies. We included 13 studies in the review and excluded 41 studies. Five out of 13 included trials did not report data that could be included in the meta-analysis. Eight studies (962 participants) provided data for meta-analysis.


9

Figure 1. Study flow diagram.


10

Included studies We included 13 studies but only eight studies (involving 962 women) contributed data for our analyses. One study compared opioid analgesics with placebo and with combination analgesia (Sunshine 1992). Nine studies assessed non-opioid analgesics versus placebo ( Angle 2002; Bjune 1996; Carvalho 2006; Fong 2008; Goheen 2000; Lee 2004; Pagnoni 1996; Sunshine 1993; Sunshine 1992). Two studies compared non-opioid and combination oral analgesics (Munishankar 2008; Sunshine 1993). Two studies assessed other oral preparations for relieving postcaesarean pain, e.g. clonidine (Palmer 2000) and gabapentin ( Short 2012). One study compared opioid and non-opioid analgesics ( Sammour 2011).Only data for adverse effects could be used from this study. Data for adequate pain relief could not be used because pain scores were described as mean +- SD of additional medication requested, they did not use visual analogue scales (VAS) even though in their methods they stated that they would use VAS. Data for breastfeeding scores could not be used because they were not as pre-specified. Data from five trials have not been included for analysis in this review due to different reasons, e.g. one trial reported only one outcome (e.g. use of morphine patient-controlled analgesia as an additional pain relief ), which was not pre-specified in our protocol (i.e. we assess need for additional analgesia (number of women) (Palmer 2000); one trial did not report the number of participants in each group; it was published as letter to the editor (Fong 2008); three trials reported outcomes that differ to the pre-specified outcomes in the review (Carvalho 2006; Goheen 2000; Munishankar 2008). Section Characteristics of included studies has the details on the above-mentioned studies.

Excluded studies We excluded 41 studies for following reasons: • four studies compared intravenous, intramuscular or subcutaneous opioids with oral non-opioids (Alhashemi 2006; Danesh 2009; Dieterich 2012; Marzida 2009); • two studies compared patient-controlled analgesia versus midwife-administered analgesia (Sibilla 1994; Snell 2006);

• two studies compared oral analgesia and epidural opioids (Choi 2003; Liu 2011); • one study compared food effect on analgesic response to non-opioids (Sunshine 1997b); • one study investigated epidural opioids compared with oral opioids for post opioids pruritus (Stuart 1995); • one study investigated pre-emptive administration of a plant extract on post-caesarean pain (Gharabaghi 2011); • one study compared combination drugs and patientcontrolled analgesia with morphine (Davis 2006); • one study investigated non-opioid analgesics to reduce opioid use post-caesarean section and had 40% of incomplete data (Esper 2005); • one study compared oral celecoxib and intravenous ketamine or their combination (Tan 2007) and this review assess only oral analgesia; • one study compared patient-controlled epidural analgesia (PCEA) and celecoxib (Nakou 2010); • one study investigated the use of celecoxib for prevention of opioid induced pruritus (Samimi 2011); • one study investigated the effectiveness of gabapentin use at delivery of neonate for pain response during intramuscular injection (Shah 2010); • one study was excluded because 28% of participants were excluded for protocol violation (Moore 2011); • one study investigated standard administered analgesics and patient-administered (East 2007) and our review does not have this subgroup; • one study assessed premedication with clonidine in reducing morphine requirements post-caesarean section (Yanagidate 2001); • one study was published as a letter to the publisher; the study was terminated after enrolling five participants because of high VAS scores (Ortner 2011); • one study assessed the effectiveness of adjunct analgesia to PCEA following a caesarean delivery. The interventions included a combination of intravenous and oral medication versus placebo and therefore the results could not be interpreted separately for oral analgesia (Paech 2014).

Risk of bias in included studies Risk of bias of included studies is presented in Figure 2 and Figure 3.


11

Figure 2. ’Risk of bias’ summary: review authors’ judgements about each risk of bias item for each included study.


12

Figure 3. ’Risk of bias’ graph: review authors’ judgements about each risk of bias item presented as percentages across all included studies.

Allocation Sequence generation was assessed as low risk of bias in eight studies, the majority of which used computer-generated random numbers (Angle 2002; Bjune 1996; Carvalho 2006; Fong 2008; Munishankar 2008; Sammour 2011; Short 2012; Sunshine 1992) and unclear risk of bias in five studies where the details of sequence generation were not described (Goheen 2000; Lee 2004; Pagnoni 1996; Palmer 2000; Sunshine 1993). Allocation concealment was assessed as low risk of bias in eight studies, with the majority of them using opaque sealed envelopes (Angle 2002; Carvalho 2006; Munishankar 2008; Pagnoni 1996; Sammour 2011; Short 2012; Sunshine 1992; Sunshine 1993). Five studies did not describe the process of allocation concealment that they used and, therefore, we considered their allocation concealment bias unclear (Bjune 1996; Fong 2008; Goheen 2000; Lee 2004; Palmer 2000).

Blinding Two studies had high risk of performance bias (Munishankar 2008; Sammour 2011). Seven studies had low risk of performance bias (Angle 2002; Carvalho 2006; Lee 2004; Pagnoni 1996; Short 2012; Sunshine 1992; Sunshine 1993) and in four studies the risk of performance was unclear (Bjune 1996; Fong 2008; Goheen 2000; Palmer 2000). Three studies had high risk of detection bias (Munishankar 2008;

Sammour 2011; Short 2012). Four studies had low risk of detection bias (Angle 2002; Carvalho 2006; Pagnoni 1996; Sunshine 1993) and in six studies the risk of detection bias was unclear (Bjune 1996; Fong 2008; Goheen 2000; Lee 2004; Palmer 2000; Sunshine 1992). Incomplete outcome data Eight studies had low risk of attrition bias (Angle 2002; Bjune 1996; Carvalho 2006; Munishankar 2008; Pagnoni 1996; Sammour 2011; Short 2012; Sunshine 1992). Four studies had high risk of attrition bias (Fong 2008; Goheen 2000; Palmer 2000; Sunshine 1993) and one study had unclear risk of attrition bias (Lee 2004). Selective reporting One study was assessed as having a low risk of reporting bias (Short 2012). However, the other 12 studies had unclear risk of reporting bias (Angle 2002; Bjune 1996; Carvalho 2006; Fong 2008; Goheen 2000; Lee 2004; Munishankar 2008; Pagnoni 1996; Palmer 2000; Sammour 2011; Sunshine 1992; Sunshine 1993). Other potential sources of bias Two studies used patient-controlled analgesia morphine in addition to the study drugs (Munishankar 2008; Palmer 2000).


13

One study used ketoprofen 100 mg intramuscularly as rescue medication and adverse events (none) reported only for the six-hour study period (Pagnoni 1996). One study used Tylenol, a combination of acetaminophen 300 mg, caffeine 15 mg and codeine 30 mg, one to two tablets every three to four hours as needed and intramuscular morphine as needed (Angle 2002). One study used Brufen 600 to 1200 mg or intramuscular ketobemidone 5 mg as rescue medication (Bjune 1996). One study used two tablets of dologesic (paracetamol and dextropoxyphene) every six hours for rescue pain relief (Lee 2004). One study used a combination of hydrocodone 5 mg/acetaminophen 500 mg or oxycodone 5 mg/acetaminophen 500 mg and intravenous morphine for resistant pain.The sample size calculation suggested that 37 women per study group were required to detect a 30% reduction in incisional pain scores on movement at 36 hours. However, due to the safety concerns with COX-2 inhibitors reported in the literature at the time, this study was terminated early (Carvalho 2006). One study was published only in a form of a letter to the editor (Fong 2008). One study was published only as an abstract (Goheen 2000). One study used parenteral morphine in the first two hours postoperatively and in the ward on request, they received oral paracetamol-dextropoxyphene; 10 women dropped out from tramadol at fixed intervals group and five dropped out from tramadol on request group, but the data were presented inclusive of these women (Sammour 2011). One study used 2 mg of intravenous morphine at five-minute intervals for the first two hours in the post-anaesthesia care unit (PACU), then 50 mg oral diclofenac every eight hours and 1 G acetaminophen every six hours for 72 hours. Breakthrough pain was managed with 2 mg subcutaneous morphine on request and subsequently the women received 10 mg oral morphine as required. This was an underpowered study (Short 2012). One study also used rescue medication (Sunshine 1992). In one study, participants received repeated doses if they complained of pain two hours after receiving the study medication (Sunshine 1993).

Effects of interventions

Need for additional pain relief with a different drug Based on one study (Sunshine 1992, involving 120 women) there was no difference in the need for additional pain relief between group of women who received tramadol and the women in the control group (risk ratio (RR) 0.33, 95% confidence interval (CI) 0.06 to 1.92, one trial, 120 women) Analysis 1.1.

Secondary outcome

Maternal drug adverse effects, e.g. nausea, vomiting, sedation, constipation, diarrhoea, drowsy, sleepy, psychological impact There was no clear difference in the rate of maternal adverse drug effects between the group of women who received tramadol versus placebo (RR 6.58, 95% CI 0.38 to 113.96, one trial, 120 women) Analysis 1.2. No data were available within the comparison of opioids and placebo for other outcomes, e.g. number of days in the hospital post-operatively, re-hospitalisation due to incisional pain, fully breastfeeding at discharge, mixed feeding at discharge, incisional pain at six weeks after caesarean section, maternal postpartum depression, effect (negative) on mother and baby interaction, cost of treatment. Subgroup analysis • High versus low dose of the same drug No difference between low (75 mg) and high (150 mg) doses of tramadol in relation to need for additional pain relief was observed RR 0.67, 95% CI 0.12 to 3.78 and RR 0.14, 95% CI 0.01 to 2.68, respectively) Analysis 8.1.

Non-opioid analgesics versus placebo - comparison 2

Primary outcomes

Adequate pain relief No data available.

Opioid analgesia versus placebo - comparison 1

Primary outcomes

Adequate pain relief No data available on this outcome.

Need for additional analgesia Six studies reported on the need for additional pain relief (Angle 2002; Bjune 1996; Lee 2004; Pagnoni 1996; Short 2012; Sunshine 1993). Overall, more women required additional analgesia (primary outcome) in the placebo group in comparison with the nonopioid analgesia group, but this difference was not statistically significant (average RR 0.70, 95% CI 0.48 to 1.01, six studies, 584 women, Tau2 = 0.18, I2 = 85%) Analysis 2.1. However, we


14

observed substantial heterogeneity, likely to be related to different non-opioid drugs (naproxen 50 mg 12 hours following caesarean section in Angle 2002, paracetamol in Bjune 1996 and Sunshine 1993, celexocib in Lee 2004, ibuprofen in Pagnoni 1996, gabapentin in Short 2012), as well as different baseline pain-relief used in different studies.

Combination oral analgesics versus placebo comparison 3

Primary outcome

Adequate pain relief Subgroup analysis • Different drugs within the same group When comparing different drugs, we observed subgroup differences (Chi2 = 27.49, df = 5 (P < 0.0001); I2 = 81.8%). Only gabapentin resulted in less need for additional pain relief versus placebo (RR 0.34, 95% CI 0.23 to 0.51, one trial, 126 women).There was no difference in the need for additional pain relief associated with the use of celecoxib (RR 0.89, 95% CI 0.59 to 1.35, one study, 60 participants), ibuprofen (RR 0.66, 95% CI 0.41 to 1.07, one study, 62 women), ketoprofen (RR 1.05, 95% CI 0.86 to 1.28, one trial, 120 women), naproxen (RR 0.11, 95% CI 0.01 to 2.00, one study, 80 women), or paracetamol (RR 0.77, 95% CI 0.43 to 1.40, two studies, 136 women) Analysis 2.1. • High versus low dose of the same drug When we compared high and low doses of the same drug, we observed a significant test for subgroup differences (Chi² = 14.06, df = 3 (P = 0.003), I² = 78.7%). Gabapentin 300 mg (RR 0.25, 95% CI 0.13 to 0.49, one study, 63 women) and 600 mg (RR 0.44, 95% CI 0.27 to 0.71, one study, 63 women) were both more effective than placebo with respect to the need for additional pain relief. Similarly, ketoprofen 100 mg resulted in fewer women needing additional pain relief (RR 0.55, 95% CI 0.39 to 0.79, one study 72 women) compared to placebo. In contrast, there was no difference between the 50 mg ketoprofen group and placebo group in terms of the number of women requiring additional pain relief (RR 0.82, 95% CI 0.64 to 1.07, one study 72 women). See Analysis 7.1.

The study (Bjune 1996) in this group did not report on this outcome.

Need for additional analgesia There was no difference in the need for additional pain relief with the use of combination analgesics versus placebo (average 0.70 ( 95% CI 0.35 to 1.40, three trials, 242 women, Tau² = 0.24, I2 = 69%) Analysis 3.1.

Subgroup analysis • Different drugs within the same group When comparing different drugs within the combination oral analgesics versus placebo comparison, we observed subgroup differences (Chi² = 5.86, df = 2. (P = 0.05); I² = 65.8%). Only paracetamol plus codeine versus placebo resulted in fewer women requiring additional pain relief (RR 0.44, 95% CI 0.23 to 0.82, one trial, 65 women) whereas there was no difference in the need for additional analgesia for the comparisons of either paracetamol plus oxycodone versus placebo or paracetamol plus propoxyphene (RR 1.00, 95% CI 0.78 to 1.28, one trial, 96 women and RR 0.65, 95% CI 0.11 to 3.69, one trial, 81 women, respectively) Analysis 3.1.

Secondary outcome Secondary outcomes

Adverse drug effects Adverse effects More side effects were reported with the use of non-opioid analgesics versus placebo (RR 11.12, 95% CI 2.13 to 58.22, two studies, 267 women) Analysis 2.2. The following outcomes were not reported in any studies: number of days in the hospital post-operatively, re-hospitalisation due to incisional pain, fully breastfeeding at discharge, mixed feeding at discharge, incisional pain at six weeks after caesarean section, maternal postpartum depression, effect (negative) on mother and baby interaction, cost of treatment.

There was an incidence of adverse drug effects in the combination oral analgesics group compared to the placebo control, but the difference was not statistically significant (RR 13.18, 95% CI 02.86 to 60.68, three studies, 252 women) Analysis 3.2. The following outcomes were not reported in any studies for this comparison: number of days in the hospital post-operatively, rehospitalisation due to incisional pain, fully breastfeeding at discharge, mixed feeding at discharge, incisional pain at six weeks after caesarean section, maternal postpartum depression, effect (negative) on mother and baby interaction, cost of treatment.


15

Opioid analgesics versus non-opioid - comparison 4

Secondary outcomes

Primary outcome

Maternal adverse effects also did not differ between the two groups (RR 6.74, 95% CI 0.39 to 116.79, one trial, 121 women) Analysis 5.2.


Need for additional analgesia There was no difference in the need for additional pain relief with a different drug in woman who received acetaminophen versus tramadol (RR 0.51, 95% CI 0.07 to 3.51, one study, 121 women) Analysis 4.1.

Secondary outcomes

Maternal drug adverse effects, e.g. nausea, vomiting, sedation, constipation, diarrhoea, drowsy, sleepy, psychological impact Adverse effects were more common with the use of combination analgesics versus placebo (RR 2.32, 95% CI 1.15 to 4.69, two trials, 241 women) Analysis 4.2. The following outcomes were not reported in any studies in this comparison: number of days in the hospital post-operatively, rehospitalisation due to incisional pain, fully breastfeeding at discharge, mixed feeding at discharge, incisional pain at six weeks after caesarean section, maternal postpartum depression, effect (negative) on mother and baby interaction, cost of treatment.

Opioid analgesics versus combination analgesics comparison 5

Primary outcome


Need for additional analgesia Based on one study comparing tramadol and paracetamol plus propoxyphene, opioid analgesics versus combination analgesics did not differ in regards to need for additional pain relief (RR 0.51, 95% CI 0.07 to 3.51, one trial, 121 women) Analysis 5.1.

Non-opioid analgesics versus combination comparison 6

Primary outcome


Need for additional pain relief with a different drug There need for additional pain relief was greater in the group of women who received non-opoid analgesics compared with the group of women who received combination analgesics (RR 0.87, 95% CI 0.81 to 0.93, one trial, 192 women) Analysis 6.1.

Secondary outcomes

No data were available for any of the secondary outcomes in this comparison.

DISCUSSION Oral analgesia is an effective and convenient form of pain relief following caesarean section. We planned to assess adequate pain relief and the need for additional analgesia as the primary outcomes to establish the efficacy of different types of oral analgesia. The included studies contributed data only on the outcome of need for additional pain relief. Although five studies reported on adequate pain relief (Angle 2002; Carvalho 2006; Goheen 2000; Pagnoni 1996; Short 2012) using various definitions for assessment of pain relief, e.g. Angle 2002 used a visual analogue scale (VAS) at rest 36 hours following caesarean section, Pagnoni 1996 included participants with a baseline VAS above 60 mm, Carvalho 2006 used a scale of 0 to 10 for VAS and we multiplied by 10, Goheen 2000 and Short 2012 used VAS at rest at 24 hours, the data could not be used as our pre-specified primary outcome measure was the adequate pain relief as reported by the woman, or by determination of more than 50% relief of pain (as either stated by the woman or calculated using a formula). None of the studies comparing non-opioids and placebo reported the data on this outcome. Only one of our pre-specified secondary outcomes were reported in the included studies, e.g. maternal drug effects. No other


16

outcomes (number of days in the hospital post-operatively, re-hospitalisation due to incisional pain, fully breastfeeding at discharge, mixed feeding at discharge, incisional pain at six weeks after caesarean section, maternal postpartum depression, effect (negative) on mother and baby interaction, cost of treatment) were reported in included trials.

Summary of main results Despite identifying a large number (63) of reports for post-caesarean pain relief, few trials (13) satisfied the inclusion criteria. Since the trials investigated different types of oral analgesia, metaanalysis was possible only on a few outcomes. Only eight small trials involving 962 women (out of 13 included trials) contributed data to the analysis, of which only four trials had low risk of bias. The evidence in most of our comparisons is based on a small number of women and the results should be interpreted with caution. No difference in the need for additional pain relief was seen between the use of opioids versus placebo, non-opioids versus placebo, combination analgesics versus placebo, opioids versus non-opioids, opioids versus combination analgesics. It is important to note the high level of heterogeneity in the comparison nonopioid analgesics versus placebo, likely to be due to different nonopioid drugs being compared (celexocib, gabapentin, ibuprofen, ketoprofen, naproxen, paracetamol). Subgroup analysis of different drugs within the same group has found that only gabapentin resulted in statistically significant decrease in the need for additional pain relief in comparison to placebo, but not other nonopioid drugs such as celexocib, ibuprofen, ketoprofen, naproxen, paracetamol. In the comparison of combination analgesia with placebo only paracetamol plus codeine (but not paracetamol plus oxycodone or plus propoxyphene) resulted in less need for additional pain relief. Maternal side effects were more common with the use of opioids, non-opioids and combination analgesics in comparison to placebo, and with opioids versus non-opioids analgesics. Subgroup analysis of different doses of the same drug found that gabapentin 300 mg and 600 mg were both more effective than placebo with respect to the need for additional pain relief. Similarly, ketoprofen 100 mg resulted in fewer women needing additional pain relief compared to placebo. In contrast, there was no difference between the 50 mg ketoprofen group, 75 mg tramadol and 150 mg tramadol groups versus placebo group in terms of the number of women requiring additional pain relief.

Overall completeness and applicability of evidence Only seven studies including 816 participants reported on the primary outcome and only two of the secondary outcomes were reported in the included trials. No data were found on the following

secondary outcomes: number of days in hospital post-operatively, re-hospitalisation due to incisional pain, fully breastfeeding on discharge, mixed feeding at discharge, incisional pain at six weeks after caesarean section, maternal post partum depression, effect (negative) on mother and baby interaction and cost of treatment. All included trials were undertaken in high-resource settings. All the trials used other forms of additional analgesia as a standard and for breakthrough pain and they all used different drugs. We have not included the data from four trials included in the review due to different reasons, e.g. one trial used patient-controlled analgesia morphine in milligrams (mg) for additional pain relief (Palmer 2000), one trial did not report the number of participants in each group (Fong 2008), two studies used a different definition of adequate pain relief to that pre-specified in the review (Goheen 2000; Sammour 2011). Three of the included studies were published in abstract form only (Fong 2008; Goheen 2000; Palmer 2000).

Quality of the evidence Eight trials with 962 participants were included in the analysis, of which only four trials (Angle 2002; Pagnoni 1996; Short 2012; Sunshine 1992) were considered to be of high quality. All the trials were small. There was high level of heterogeneity for need of additional pain relief in a comparison between non-opioid oral analgesia and placebo, due to various drugs with different mechanisms of action combined in the analysis (naproxen, ibuprofen, paracetamol, celexocib, gabapentin). Although the need for additional pain relief was more common with the use of naproxen, ibuprofen, paracetamol, celecoxib and gabapentin, this difference reached statistical significance only with the use of gabapentin. There was a low level of missing data in the included trials except for one trial comparing combination oral analgesia versus placebo (Bjune 1996) with 16% of participants excluded for protocol violation.

Potential biases in the review process We have performed meta-analysis of studies using different forms of baseline pain relief without taking into consideration the differences. We carried out subgroup analysis for different drugs within the same group and for high versus low doses of the same drug. However, with relatively few studies (one to two trials) and numbers of participating women (40 to 136) this limits the reliability of these subgroup analyses.

Agreements and disagreements with other studies or reviews We did not identify any other meta-analysis assessing oral analgesia for post-caesarean pain. There is a large number of studies assessing


17

other modalities of pain relief following caesarean section, e.g. systemic (intravenous and intramuscular analgesics, nerve block techniques, wound infiltration (Ismail 2012). Although the results of this review should be interpreted with caution due to small numbers and other shortcomings, the absence of efficacy of oral analgesics in comparison to placebo may suggest the need for other forms of analgesia to be used for post-caesarean pain relief.

AUTHORS’ CONCLUSIONS Implications for practice No conclusions on the best available option for oral analgesia for pain relief post-caesarean section can be drawn from this review. Although the results of this review should be interpreted with caution due to the small numbers and other shortcomings, the absence of efficacy of oral analgesics in comparison to placebo may suggest the need for other forms of analgesia to be used for postcaesarean pain relief. Further well-designed studies are necessary to establish the safe and effective form of oral analgesia for postcaesarean pain.

Implications for research Because of the risk of bias and small numbers in the studies re-

viewed, adequately sized and well-designed trials are needed to assess different types of oral analgesia for post-caesarean pain. Further research is needed to compare opioid analgesics versus placebo/no drug treatment, non-opioid analgesics versus placebo/ no drug treatment, combination drugs verus placebo/no drug treatment, opioid analgesics versus non-opioid analgesics, opioid analgesics versus combination analgesics, non-opioid analgesics versus combination analgesia in relation to pain relief. In addition, other important outcomes should be considered, e.g. side effects, number of days in the hospital post-operatively, incisional pain, breastfeeding, maternal postpartum depression, effect on mother and baby interaction and cost of treatment comparing in group of women who primiparous and multiparous, undergoing elective and emergency caesarean section, with and without baseline pain relief.

ACKNOWLEDGEMENTS Cochrane Pregnancy and Childbirth editorial team for administrative and technical support. As part of the pre-publication editorial process, this review has been commented on by three peers (an editor and two referees who are external to the editorial team) and the Group’s Statistical Adviser.

REFERENCES

References to studies included in this review Angle 2002 {published data only} Angle PJ, Halpern S, Leighton B, Wilson DS, Gnanendran K, Kronberg J. Naproxen improves post cesarean analgesia after spinal morphine [abstract]. Anesthesia & Analgesia 2000;90 Suppl:Abstract no: 281. ∗ Angle PJ, Halpern SH, Leighton BL, Szalai JP, Gnanendran K, Kronberg JE. A randomized controlled trial examining the effect of naproxen on analgesia during the second day after cesarean delivery. Anesthesia & Analgesia 2002;95:741–9. Bjune 1996 {published data only} Bjune K, Stubhaug A, Dodgson MS, Breivik H. Additive analgesic effect of codeine and paracetamol can be detected in strong, but not moderate, pain after caesarean section Baseline pain-intensity is a determinant of assay-sensitivity in a postoperative analgesic trial [see comments]. Acta Anaesthesiologica Scandinavica 1996;40(4):399–407. Carvalho 2006 {published data only} ∗ Carvalho B, Chu L, Fuller A, Cohen SE, Riley ET. Valdecoxib for postoperative pain management after cesarean delivery: a randomized, double-blind, placebo-

controlled study. Anesthesia & Analgesia 2006;103(3): 664–70. Fuller AJ, Brummel C, Saxena A, Chu L, Riley ET, Cohen SE, et al. Valdecoxib for postoperative pain management after cesarean section: a randomized, double-blinded, placebo-controlled study [abstract]. Anesthesiology 2005; 102(Suppl 1):15. Fong 2008 {published data only} Fong WP, Yang LC, Wu JI, Chen HS, Tan PH. Does celecoxib have pre-emptive analgesic effect after caesarean section surgery?. British Journal of Anaesthesia 2008;100(6): 861–2. Goheen 2000 {published data only} Goheen MSLB, Ong BY, Wahba RF, Lucy SJ. Effect of oral naproxen on pain following cesarean section [abstract]. Anesthesiology 2000;92 Suppl:Abstract no: A39. Lee 2004 {published data only} Lee LHY, Irwin MG, Lim J, Wong CK. The effect of celecoxib on intrathecal morphine-induced pruritis in patients undergoing caesarean section. Anaesthesia 2004;59: 876–80. Munishankar 2008 {published data only} Munishankar B, Fettes P, Moore C, McLeod GA. A double-


18

blind randomised controlled trial of paracetamol, diclofenac or the combination for pain relief after caesarean section. International Journal of Obstetric Anesthesia 2008;17(1): 9–14. Pagnoni 1996 {published data only} Pagnoni B, Vignali M, Colella S, Monopoli R, Tiengo N. Comparative efficacy of oral ibuprofen arginine and intramuscular ketorolac in patients with postcaesarean section pain. Clinical Drug Investigation 1996;11:15–21. Palmer 2000 {published data only} Palmer CM, Nogami W, Alves D. Oral clonidine: use with intrathecal morphine for post-cesarean analgesia [abstract]. Anesthesiology 2000;92 Suppl:Abstract no: A57. Sammour 2011 {published data only} Sammour R, Ohel G, Cohen M, Gonen R. Tramadol versus naproxen for post-cesarean section analgesia - a randomized trial. American Journal of Obstetrics and Gynecology 2009; 201(6 Suppl 1):S63. ∗ Sammour RN, Ohel G, Cohen M, Gonen R. Oral naproxen versus oral tramadol for analgesia after cesarean delivery. International Journal of Gynecology & Obstetrics 2011;113(2):144–7. Short 2012 {published data only} Short J, Downey K, Bernstein P, Shah V, Carvalho JCA. A single preoperative dose of gabapentin does not improve postcesarean delivery pain management: A randomized, double-blind, placebo-controlled dose-finding trial. Anesthesia & Analgesia 2012;115(6):1336–42. Sunshine 1992 {published data only} Sunshine A, Olson NZ, Zighelboim I, DeCastro A, Minn FL. Analgesic oral efficacy of tramadol hydrochloride in postoperative pain. Clinical Pharmacology and Therapeutics 1992;51(6):740–6. Sunshine 1993 {published data only} Sunshine A, Olson NZ, Zighelboim I, De Castro A. Ketoprofen, acetaminophen plus oxycodone, and acetaminophen in the relief of postoperative pain. Clinical Pharmacology and Therapeutics 1993;54:546–55.

References to studies excluded from this review Abboud 1990 {published data only} Abboud TK, Afrasiabi A, Davidson J, Zhu J, Reyes A, Khoo N, et al. Prophylactic oral naltrexone with epidural morphine: effect on adverse reactions and ventilatory responses to carbon dioxide. Anesthesiology 1990;72:233–7. Abboud TK, Lee K, Chai M, Zhu J, Afrasiabi A, Mantilla M, et al. Prophylactic oral naltrexone with intrathecal morphine for cesarean section: effects on adverse reactions and analgesia. Anesthesiology 1989;71:Abstract no: A836. ∗ Abboud TK, Lee K, Zhu J, Reyes A, Afrasiabi A, Mantilla M, et al. Prophylactic oral naltrexone with intrathecal morphine for cesarean section: effects on adverse reactions and analgesia. Anesthesia and Analgesia 1990;71:367–70. Alhashemi 2006 {published data only} Alhashemi JA, Alotaibi QA, Mashaat MS, Kaid TM, Mujallid RH, Kaki AM. Intravenous acetaminophen vs

oral ibuprofen in combination with morphine PCIA after cesarean delivery. Canadian Journal of Anesthesia 2006;53 (12):1200–6. Bloomfield 1993 {published data only} Bloomfield SS, Cissell GB, Mitchell J, Barden TP, Kaiko RF, Fitzmartin RD, et al. Analgesic efficacy and potency of two oral controlled-release morphine preparations. Clinical Pharmacology & Therapeutics 1993;53(4):469–78. Carvalho 2007 {published data only} Carvalho B, Roland LM, Chu LF, Campitelli III VA, Riley ET. Single-dose, extended-release epidural morphine (DepoDur) compared to conventional epidural morphine for post-cesarean pain. Anesthesia & Analgesia 2007;105(1): 176–83. Carvalho 2008 {published data only} Carvalho JCA. The effectiveness and safety of gabapentin for post-operative pain after cesarean section. ClinicalTrials.gov (http://clinicaltrials.gov/) (accessed 20 February 2008). NCT00573664 2008. Choi 2003 {published data only} ∗ Choi DM, Kliffer AP, Douglas MJ. Dextromethorphan and intrathecal morphine for analgesia after caesarean section under spinal anaesthesia. BJA: British Journal of Anaesthesia 2003;90(5):653–8. Choi DMA, Kliffer AP, Douglas MJ. Oral dextromethorphan and intrathecal morphine for analgesia after caesarean section [abstract]. Anesthesiology 2000;92 Suppl:Abstract no: A30. Danesh 2009 {published data only} Danesh A, Jabal Ameli M, Ghaedi S. Comparison the efficacy of oral methadone and IM pethidine for pain relief after cesarean section. Regional Anesthesia and Pain Medicine 2009;34(5):150. Davis 2006 {published data only} Davis KM, Esposito MA, Meyer BA. Oral analgesia compared with intravenous patient-controlled analgesia for pain after cesarean delivery: a randomized controlled trial. American Journal of Obstetrics and Gynecology 2006;194(4): 967–71. De Lia 1986 {published data only} De Lia JE, Rodman KC, Jolles CJ. Comparative efficacy of oral flurbiprofen, intramuscular morphine sulfate, and placebo in the treatment of gynecologic postoperative pain. American Journal of Medicine 1986;80(3A):60–4. Dieterich 2012 {published data only} Dieterich M, Muller-Jordan K, Stubert J, Kundt G, Wagner K, Gerber B. Pain management after cesarean: a randomized controlled trial of oxycodone versus intravenous piritramide. Archives of Gynecology and Obstetrics 2012;286(4):859–65. East 2007 {published data only} East N, Dube J, Perreault EL. Postpartum pain relief: a randomized comparison of self-administered medication and standard administration. Journal of Obstetrics and Gynaecology Canada 2007;29(12):975–81.


19

Esper 2005 {published data only} Esper D, Boggess K, Potter C, Arnette R, Mayer D. A pilot study of valdecoxib for reduction of postoperative opioid use after cesarean delivery [abstract]. American Journal of Obstetrics and Gynecology 2005;193(6 Suppl):S124.

Ortner 2011 {published data only} Ortner CM. Patient-controlled oral analgesia following cesarean section: tramadol versus a combination of tramadol and acetaminophen. Acta Obstetricia et Gynecologica Scandinavica 2011;90(8):925–6.

Ganem 2003 {published data only} Ganem EM, Modolo NSP, Ferrari F, Cordon FCO, Koguti ES, Castiglia YMM. Effects of low spinal morphine doses associated to intravenous and oral ketoprofen in patients submitted to cesarean sections [Efeitos da associacao entre pequenas doses subaracnoideas de morfina e cetoprofeno venoso e oral em pacientes submetidas a cesariana]. Revista Brasileira de Anestesiologia 2003;53(4):431–9.

Paech 2014 {published data only} Paech MJ, McDonnell NJ, Sinha A, Baber C, Nathan EA. A randomised controlled trial of parecoxib, celecoxib and paracetamol as adjuncts to patient-controlled epidural analgesia after caesarean delivery. Anaesthesia & Intensive Care 2014;42(1):15–22.

Gharabaghi 2011 {published data only} Gharabaghi PM, Tabatabei F, Fard SA, Sayyah-Melli M, Ouladesahebmadarek E, Del Azar A, et al. Evaluation of the effect of preemptive administration of Rosa damascena extract on post-operative pain in elective cesarean sections. African Journal of Pharmacy and Pharmacology 2011;5(16): 1950–5. Liu 2011 {published data only} Liu Q, Zhao JN, Chen HX, Zhong TD. [Oral controlledrelease oxycodone for uterine cramping pain after cesarean section]. [Chinese]. Chung-Hua i Hsueh Tsa Chih [Chinese Medical Journal] 2011;91(30):2132–4. Marzida 2009 {published data only} Marzida M. A randomized controlled study comparing subcutaneous pethidine with oral diclofenac for pain relief after caesarean section. Journal of the University of Malaya Medical Centre 2009;12(2):63–9. McDonnell 2010 {published data only} McDonnell NJ, Paech MJ, Browning RM, Nathan EA. A randomised comparison of regular oral oxycodone and intrathecal morphine for post-caesarean analgesia. International Journal of Obstetric Anesthesia 2010; 19(1): 16–23. Moore 2011 {published data only} Moore A, Costello J, Wieczorek P, Shah V, Taddio A, Carvalho JC. Gabapentin improves postcesarean delivery pain management: a randomized, placebo-controlled trial. Anesthesia & Analgesia 2011;112(1):167–73. Morrison 1986 {published data only} Morrison JC, Harris J, Sherrill J, Heilman CJ, Bucovaz E T, Wiser WL. Comparative study of flurbiprofen and morphine for postsurgical gynecologic pain. American Journal of Medicine 1986;80(3A):55–9. Nakou 2010 {published data only} Nakou M, Matsota P, Kalimeris K, Batistaki C, Pandazi A, Kostopanagiotou. A single dose of celecoxib improves pain scores and quality of patient controlled epidural analgesia after caesarean section. Regional Anesthesia and Pain Medicine 2010;35(5):E142–3. Olson 1984 {published data only} Olson N, Sunshine A, Roure C, Colon A, Laska EM, Santiago H, et al. Analgesic efficacy of suprofen, codeine and placebo. Pain 1984;2 Suppl:238.

Palangio 2000 {published data only} Palangio M, Wideman GL, Keffer M, Landau CJ, Morris E, Doyle RTJ, et al. Combination hydrocodone and ibuprofen versus combination oxycodone and acetaminophen in the treatment of postoperative obstetric or gynecologic pain. Clinical Therapeutics 2000;22:600–12. Samimi 2011 {published data only} ∗ Samimi S, Davari Tanha F, Malekian M. A blinded study using celecoxib for prevention of morphine induced pruritus in patients undergoing cesarean section. Journal of Family and Reproductive Health 2011;5(2):35–9. Samimi Sadeh S, Davari F, Malekian M. A blinded study using celecoxib for prevention of morphine induced pruritus in patients undergoing cesarean section. Regional Anesthesia and Pain Medicine 2012;37(5 Suppl 1):E307. Shah 2010 {published data only} Shah VS, Gerges S, Moore A, Carvalho J, Taddio A. The effectiveness of maternal gabapentin therapy at the time of delivery on neonatal pain response during intramuscular injection. Pediatric Academic Societies’ 2010 Annual Meeting; 2010 May 1-4; Vancouver, Canada. 2010. Sibilla 1994 {published data only} Sibilla C, Albertazzi P, Zatelli R, Lupi G, Marchi M, Campobasso C, et al. Pain relief after caesarean section: comparison of different techniques of morphine administration. International Journal of Obstetric Anesthesia 1994;3:203–7. Snell 2006 {published data only} Snell P, Hicks C. An exploratory study in the UK of the effectiveness of three different pain management regimens for post-caesarean section women. Midwifery 2006;22: 249–61. Stuart 1995 {published data only} Stuart AL, Carter JB, Roberson CR, Morton GH, O’Rear JA, Dutton DA. Epidural butorphanol and oral naltrexone reduce pruritus from epidural morphine after Cesarean birth. Anesthesia and Analgesia 1995;80:S479. Sunshine 1983 {published data only} Sunshine A, Olson JZ, Siegel C, Laska EM. Oral analgesic study of ketoprofen, aspirin and placebo in post-partum pain. Clinical Pharmacology and Therapeutics 1983;33:154. Sunshine 1985a {published data only} Sunshine A, Roure C, Zorrilla C, Laska EM, Olson N, Rivera J. The analgesic efficacy of ibuprofen, codeine and


20

placebo. Clinical Pharmacology and Therapeutics 1985;37: 232. Sunshine 1985b {published data only} Sunshine A, Zighelboim I, Olson NZ, De Sarrazin C, Laska E. A comparative oral analgesic study of indoprofen, aspirin, and placebo in postpartum pain. Journal of Clinical Pharmacology 1985;25(5):374–80. Sunshine 1986 {published data only} Sunshine A, Zighelboim I, Laska E, Siegel C, Olson NZ, De Castro A. A double-blind, parallel comparison of ketoprofen, aspirin, and placebo in patients with postpartum pain. Journal of Clinical Pharmacology 1986;26 (8):706–11. Sunshine 1987 {published data only} Sunshine A, Roure C, Olson N, Laska EM, Zorrilla C, Rivera J. Analgesic efficacy of two ibuprofen-codeine combinations for the treatment of postepisiotomy and postoperative pain. Clinical Pharmacology and Therapeutics 1987;42:374–80. Sunshine 1988 {published data only} Sunshine A, Axtmayer R, Olson NZ, Laska E, Ramos I. Analgesic efficacy of pentazocine versus a pentazocinenaloxone combination following oral administration. Clinical Journal of Pain 1988;4(1):35–40. Sunshine 1989 {published data only} Sunshine A, Laska E, Siegel C, Zighelboim I, De Castro A, Sorrentino J, et al. Analgesic adjuvancy of caffeine with ibuprofen in three different postpartum pain populations. Clinical Pharmacology and Therapeutics 1989;45:174. Sunshine 1997a {published data only} Sunshine A, Olson NZ, O’Neill E, Ramos I, Doyle R. Analgesic efficacy of a hydrocodone with ibuprofen combination compared with ibuprofen alone for the treatment of acute postoperative pain. Journal of Clinical Pharmacology 1997;37:908–15. Sunshine 1997b {published data only} ∗ Sunshine A, Olsen NZ, Zighelboim I, Wajdula JS. A food interaction study of bromfenac, naproxen sodium and placebo in caesarean section patients. Clinical Pharmacology and Therapeutics 1997;61:PIII–9. Sunshine A, Olson NZ, Zighelboim I, Wajdula JS. A food interaction study of bromfenac, naproxen sodium, and placebo in cesarean section patients [abstract]. Primary Care Update for Ob/Gyns 1998;5(4):195. Tan 2007 {published data only} Tan PH, Wu JI, Yang LC. Analgesic effect of pre-incisional celecoxib, ketamine and their combination for caesarean section [abstract]. Regional Anesthesia and Pain Medicine 2007;32(5 Suppl):81. Wittels 1993 {published data only} Wittels B, Glosten B, Faure E, Moawad A, Ismail M, Hibbard J, et al. Opioid antagonist adjuncts to epidural morphine for postcesarean analgesia: maternal outcomes. Anesthesiology 1992;77:A1013. ∗ Wittels B, Glosten B, Faure EAM, Moawad AH, Ismail M, Hibbard J, et al. Opioid antagonist adjuncts to epidural

morphine for postcesarean analgesia: maternal outcomes. Anesthesia and Analgesia 1993;77:925–32. Yanagidate 2001 {published data only} Yanagidate F, Hamaya Y, Dohi S. Clonidine premedication reduces maternal requirement for intravenous morphine after cesarean delivery without affecting newborn’s outcome. Regional Anesthesia and Pain Medicine 2001;26(5):461–7.

Additional references Berghella 2012 Berghella V. Caesarean delivery: postoperative issues. http://www.uptodate.com/contents/cesarean-deliverypostoperative-issues (accessed 2013) 2012. Betran 2007 Betrán AP, Merialdi M, Lauer JA, Bing-Shun W, Thomas J, Van Look P, et al. Rates of caesarean section: analysis of global, regional and national estimates. Paediatric and Perinatal Epidemiology 2007;21(2):98–113. Chahl 1996 Chahl LA. Opioids - mechanisms of action. Australian Prescriber 1996;19:63–5. Chandrasekharan 2002 Chandrasekharan NV, Dai HK, Roos KL, Evanson NK, Tomsik J, Elton TS. COX-3, a cyclooxygenase-1variant inhibited by acetaminophen and other analgesic/antipyretic drugs: cloning, structure, and expression. Proceedings of the National Academy of Sciences of the United States of America 2002;99(21):926-31. Chou 2009 Chou D, Abalos E, Gyte GMI, Gulmezoglu AM. Drugs for perineal pain in the early postpartum period: generic protocol. Cochrane Database of Systematic Reviews 2009, Issue 3. [DOI: 10.1002/14651858.CD007734.pub2] Cooper 1991 Cooper SA. Commentary: single-dose analgesic studies: the upside and downside of assay sensitivity. In: Max M, Portenov R, Laska E editor(s). Advances in Pain Research and Therapy. Vol. 18, New York: Raven Press, 1991:117–24. Delbaere 2012 Delbaere I, Cammu H, Martens E, Tency I, Martens G, Temmerman M. Limiting the caesarean section rate in low risk pregnancies is key to lowering the trend of increased abdominal deliveries: an observational study. BMC Pregnancy Childbirth 2012;12:3. [DOI: 10.1186/ 1471-2393-12-3] DiMatteo 1996 DiMatteo MR, Morton SC, Lepper HS, Damush TM, Carney MF, Pearson M, et al. Cesarean childbirth and psychosocial outcomes: a meta-analysis. Health Psychology 1996;15(4):303–14. Gadsden 2005 Gadsden J, Hart S, Santos AC. Post-ceaserean delivery analgesia. Anesthesia & Analgesia 2005;101:S62–S69.


21

Getahun 2009 Getahun D, Strickland D, Lawrence J, Fassett M, Koebnick C, Jacobsen S. Racial and ethnic disparities in the trends in primary cesarean delivery based on indications. American Journal of Obstetrics & Gynecology 2009;201(422):e1–7. [DOI: 10.1016/j.ajog.2009.07.062] Gibbons 2010 Gibbons L, Belizán J, Lauer J, Betrán A, Merialdi M, Althabe F. The Global Numbers and Costs of Additionally Needed and Unnecessary Caesarean Sections Performed per Year: Overuse as a Barrier to Universal Coverage: World Health Report. Geneva: WHO, 2010. Gordh 1989 Gordh T Jr, Jansson I, Hartvig P, Gillberg PG, Post C. Interactions between noradrenergic and cholinergic mechanisms involved in spinal nociceptive processing. Acta Anaesthesiologica Scandinavica 1989;33(1):39–47. Hidalgo 2005 Hidalgo MP, Auzani JA, Rumpel LC. The clinical effect of small oral clonidine doses on perioperative outcomes in patients undergoing abdominal hysterectomy. Anesthesia and Analgesia 2005;100(3):795–802. Higgins 2011 Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from www.cochrane-handbook.org. Ingrid 2006 Ingrid M. Does caesarean section negatively influence the post partum prognosis of low back pain and pelvic pain during pregnancy?. European Spine Journal 2007;16(1): 115–21. Ismail 2012 Ismail S. What is new in postoperative analgesia after caesarean sections?. Anaesthesia, Pain & Intensive Care 2012; 16(2):123–6. Kainu 2010 Kainu JP, Sarvela J, Tiippana E, Halmesmäki E, Korttila KT. Persistent pain after caesarean section and vaginal birth: a cohort study. International Journal of Obstetric Anesthesia 2010;19(1):4–9. Kelly 2001 Kelly DJ, Ahmad M, Brull SJ. Preemptive analgesia I: physiological pathways and pharmacological modalities. Canadian Journal of Anesthesia 2001;48(10):1000–10. Kodali 2012 Kodali B-S, Oberoi JS. Management of post operative pain. http://www.uptodate.com/contents/management-ofpostoperative-pain (accessed 2013) 2012. Kuo 2006 Kuo G. Nonsteroidal anti-inflammatory drugs. In: BoswellM, Cole BE editor(s). Weiner’s Pain Management: a Practical Guide for Clinicians. 8th Edition. CRC Press: Boca Raton, 2006:773–88.

MacDorman 2008 MacDorman MF, Menacker F, Declercq E. Cesarean birth in the United States: epidemiology, trends and outcomes. Clinics in Perinatolology June 2008;35(2):293–303. Mathiesen 2007 Mathiesen O, Møiniche S, Dahl JB. Gabapentin and postoperative pain: a qualitative and quantitative systematic review, with focus on procedure. BMC Anesthesiology 2007; 7(7):6. [DOI: 10.1186/1471-2253-7-6] Moore 1967 Moore A, McQuay H, Gavaghan D. Deriving dichotomous outcome measures from continuous data in randomised controlled trials of analgesia: verification from independent data. Pain 1967;69(1-2):127–30. Moore 1967a Moore A, Moore O, McQuary H, Gavaghan D. Deriving dichotomous outcome measures from continuous data in randomised controlled trials of analgesia: use of pain intensity and visual analogue scales. Pain 1967;69(3): 311–5. Moore 2009 Moore RA, Straube S, Wiffen PJ, Derry S, McQuay HJ. Pregabalin for acute and chronic pain in adults. Cochrane Database of Systematic Reviews 2009, Issue 3. [DOI: 10.1002/14651858.CD007076.pub2] RevMan 2014 [Computer program] The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan). Version 5.3. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014. SAMF 2012 Anon. Analgesics. In: Rossiter D, Blockman M editor (s). South African Medicines Formulary. 10th Edition. Cape Town: Health and Medical Publishing Group, 2012: 429–38. Shang 2003 Shang AB, Gan TJ. Optimising postoperative pain management in the ambulatory patient. Drugs 2003;63(9): 855–67. Solomon 2012 Solomon DH. NSAIDs: mechanism of action. http:// www.uptodate.com/contents/nsaids-mechanism-of-action (accessed 2013) 2012. Sung 2000 Sung CS, Lin SH, Chan KH. Effect of oral clonidine premedication on perioperative hemodynamic response and postoperative analgesic requirement for patients undergoing laparoscopic cholecystectomy. Acta Anaesthesiologica Sinica 2000;38(1):23–9. Woolf 1993 Woolf CJ, Chong MS. Preemptive analgesia--treating postoperative pain by preventing the establishment of central sensitization. Anesthesia & Analgesia 1993;77(2): 362–79. ∗ Indicates the major publication for the study


22

CHARACTERISTICS OF STUDIES

Characteristics of included studies [ordered by study ID] Angle 2002 Methods

Randomised controlled trial, two arms (placebo and naproxen)

Participants

80 ASA I and II women scheduled for elective caesarean delivery were recruited into the study. Exclusion criteria included contraindication to spinal anaesthesia, diabetes, and pre-eclampsia

Interventions

Group 1 (40 women): placebo group - received Anusol suppository and there after oral lactose capsules every 12 hours for 6 doses Group 2 (40 women): Naproxen sodium 500 mg suppository followed by oral naproxen sodium 550 mg every 12 hours for 6 doses

Outcomes

Primary outcomes - incisional pain on sitting in 48 hours. Secondary outcomes - VAS pain scores, analgesic requirements, quality of analgesia and side effects

Notes Risk of bias Bias

Authors’ judgement

Support for judgement

Random sequence generation (selection Low risk bias)

Computer-generated randomisation tables (block randomised in random groups of 4 and 8). Pharmacy staff maintained randomisation key until completion of the study)

Allocation concealment (selection bias)

Low risk

Randomisation codes concealed in sequentially ordered sealed opaque envelopes

Blinding of participants and personnel Low risk (performance bias) All outcomes

Staff, participants and interviewers were blinded to treatment groups. Pharmacy supplied identical drug forms for administration

Blinding of outcome assessment (detection Low risk bias) All outcomes

The interviewers were blinded.

Incomplete outcome data (attrition bias) All outcomes

Low risk

No missing data reported.

Selective reporting (reporting bias)

Unclear risk

No prior registration of the trial protocol.


23

Angle 2002

(Continued)

Other bias

Unclear risk

This study used Tylenol, a combination of acetaminophen 300 mg, caffeine 15 mg and codeine 30 mg, 1 to 2 tablets every 3 to 4 hours as needed and intramuscular morphine as needed as a baseline analgesia

Bjune 1996 Methods

Randomised controlled trial, three arms.

Participants

125 women with strong or moderate pain according to the VAS were enrolled to the study after informed consent was obtained at day 1 or 2 post-caesarean section. Excluded were women with severe pre-eclampsia and those with children in the NICU

Interventions

Paracetamol 800 mg + codeine 60 mg (50 women), paracetamol 1000 mg (50 women) and placebo (25 women) Ibuprofen 600-1200 mg or Ketobemidone 5 mg intramuscularly were used as rescue analgesics. Rescue medication was given at least 1 hour after intake of the study drug, and then pain intensity at time of rescue analgesic intake was used as an estimate of the pain intensity for the remaining 6-hour observations

Outcomes

Satisfaction with pain, side effects.

Notes

17 (16%) women were excluded for protocol violation, 108 women were included in the analysis of need for additional analgesia and all 125 in the data for adverse effects

Risk of bias Bias




125 women were randomised into blocks of 25 and 10 to receive paracetamol 800 mg plus codeine 60 mg and paracetamol 1 g, respectively and 5 to receive placebo


Not described.

Unclear risk

Blinding of participants and personnel Unclear risk (performance bias) All outcomes

Double-blinded, however, no details were given.

Blinding of outcome assessment (detection Unclear risk bias) All outcomes

Not specified.


No reported incomplete data outcome.

Low risk


24

Bjune 1996

(Continued)


Unclear risk

No prior trial registration.

Other bias

High risk

This study used Brufen 600-1200 mg or intramuscular ketobemidone 5 mg as rescue medication. The same number of women reported on side effects in both intervention groups. 17 of 125 women were excluded due to protocol violation before the opening of randomisation code and analysis of the effect on data

Carvalho 2006 Methods

Randomised controlled trial, two groups.

Participants

48 ASA I and II women undergoing an elective caesarean section under spinal anaesthesia were enrolled into the study. 1 woman was withdrawn by her obstetrician who desired to give her NSAIDs

Interventions

Group 1 (23 women): oral valdecoxib 20 mg given 90 minutes after spinal block and every 12 hours for a total of 6 doses over 72 hours Group 2 (25 women): placebo. Hydrocodone 5 mg and acetaminophen 500 mg, oxycodone 5 mg and acetaminophen 500 mg and IV morphine for severe or resistant pain

Outcomes

Pain intensity ratings using VPS scores, post-operative analgesic consumption Secondary outcomes included: time to first analgesia, women requiring post-operative IV medication

Notes

The results presented in this paper could not be used in the meta-analysis due to difference in the outcomes reported by authors and pre-specified in the review

Risk of bias Bias




Computer-generated sequence was used.


Opaque envelopes were used.

Low risk


Double blinding. The intervention and placebo looked the same


The assessor was blinded to the intervention.


25

Carvalho 2006

(Continued)


Low risk

1 patient was withdrawn by her obstetrician due to desire of administrating NSAIDs, but the data for this participant were included as per intention-to-treat analysis


Unclear risk

No pre-trial registration of protocol.

Other bias

High risk

This study used a combination of hydrocodone 5 mg/acetaminophen 500 mg or oxycodone 5 mg/acetaminophen 500 mg and IV morphine for resistant pain The sample size calculation suggested that 37 women per study group were required to detect a 30% reduction in incisional pain scores on movement at 36 hours. However, due to the safety concerns with COX-2 inhibitors reported in literature at the time, this study was terminated early

Fong 2008 Methods

Randomised controlled trial, three groups.

Participants

60 women undergoing caesarean delivery under spinal anaesthesia

Interventions

Pre-operative group received celecoxib 400 mg 30 minutes before anaesthesia and a placebo tablet after wound closure Post-operative group received a placebo tablet 30 minutes before anaesthesia and celecoxib 400 mg after wound closure Control group received placebo 30 minutes before anaesthesia and after wound closure Number of women in each group was not reported.

Outcomes

Total morphine consumption, time to first analgesic demand, side effects

Notes

It is not stated how many were allocated in each group and therefore data cannot be used for meta-analysis. This study was published only as a letter to the editor

Risk of bias Bias




Allocation to groups by random numbers.


Not described.

Unclear risk


26

Fong 2008

(Continued)


Not described except double blind.


Not described-details about analysis not explained.


High risk

The number of participants in each group is not stated. The number of outcomes reported in each groups is not reported


Unclear risk

No prior protocol registration.

Other bias

High risk

This study was published only in a form of a letter to the editor

Goheen 2000 Methods


Participants

102 women undergoing caesarean section under spinal anaesthesia (bupivacaine 0.75%, fentanyl 15 mcg, morphine 150 mcg)

Interventions

Each group received 2 pills in PACU and then 1 pill every 8 hours. The groups were naproxen at PACU and naproxen 250 mg every 8 hours in the ward (35 women), placebo at PACU and placebo in the ward (33 women) or naproxen at PACU and placebo in the ward (34 women) Additional analgesic medication was used in women requiring that and it not specified

Outcomes

VAS scores 4 hours post spinal injection, post-operative days 1, 2 and 3 at rest and ambulating

Notes

No results from this study contributed to the meta-analysis due to a differences in presented outcomes. The study was published in abstract form only

Risk of bias Bias



Random sequence generation (selection Unclear risk bias)

No description of sequence generation.


No description of allocation concealment.

Unclear risk

Blinding of participants and personnel Unclear risk (performance bias) All outcomes Oral analgesia for relieving post-caesarean pain (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

Not described, except naproxen and placebo were visually identical 27

Goheen 2000

(Continued)


Not described.


High risk

110 women were randomised, however, “the records of 102 women were eligible for data analysis”. The authors did not state why the other records were not eligible


Unclear risk

No prior protocol registration.

Other bias

Unclear risk

This study was published only as an abstract.

Lee 2004 Methods

Randomised double-blinded trial, two arms.

Participants

60 women who are class ASA 1 or 2 who are undergoing a caesarean delivery under spinal anaesthesia

Interventions

Group 1 (30 women): celecoxib 200 mg after delivery of the baby Group 2 (30 women): placebo.

Outcomes

Need for additional pain relief.





Random number allocation, the details not described.


Not described.

Unclear risk


Capsule of celecoxib was used as a study drug and empty capsule was used as placebo. Double-blinded, e.g. the women and the anaesthetist were blinded to the intervention. The intervention and the placebo looked identical


Not clearly stated.


28

Lee 2004

(Continued)


Unclear risk

Not noted.


Unclear risk

No prior registration of protocol.

Other bias

Unclear risk

Study used 2 tablets of dologesic (paracetamol and dextropoxyphene) every 6 hours for rescue pain relief. The aim of this study was to assess the effect of celexocib on intrathecal morphine induced pruritis in women undergoing elective caesarean section

Munishankar 2008 Methods


Participants

78 women undergoing an elective caesarean section between ages 18-45 years were registered for the study. Exclusion criteria included significant maternal medical history, obstetric illness and any evidence of fetal compromise

Interventions

Group 1 (26 women): paracetamol 1 G suppository then 1 G orally 6 hourly and placebo 8 hourly Group 2 (26 women): diclofenac 100 mg suppository then 50 mg orally, 8 hourly and placebo 6 hourly Group 3 (26 women): received, paracetamol 1 G and diclofenac 100 mg suppository, then diclofenac 50 mg 8 hourly and paracetamol 1 G 6 hourly All women had caesarean section under spinal anaesthesia with 5 mg/mL of bupivacaine with dextrose 80 mg/mL in a volume of 2.25 to 2.5 mL for those < 170 cm and 2.5 to 2.75 mL for those > 170 cm mixed with fentanyl 12.5 mcg PCA morphine was given to all participants at 1 mg every 5 minutes

Outcomes

Pain at rest and on movement, side effects such as nausea and vomiting, time taken from end of surgery to drink, defecate, mobilise and overall patient satisfaction





Computer-generated random numbers.


Randomisation number were kept in sealed envelopes in the hospital pharmacy till the end of the day

Low risk


29

Munishankar 2008

(Continued)

Blinding of participants and personnel High risk (performance bias) All outcomes

Double-blinded. At the end of the surgery, the sealed envelope containing the drug allocation information was handed over to the surgical nurse who administered the medication. This nurse had no further involvement with the patient and with the study. The oral study drug were distributed to the wards in 2 plastic dispensing bags (1 bag contained 4 tablets of either paracetamol or placebo and the other one 3 tablets of either diclofenac or placebo). We consider that there was a risk of disclosure of group allocations by the surgical nurse as well as due to differences in number of tablets administered to each group as well as the different time of administration (8 hourly diclofenac group and 6 hourly paracetamol group)

Blinding of outcome assessment (detection High risk bias) All outcomes

Person who administered medication was blinded. See justification above


Low risk

No missing data.


Unclear risk


Other bias

Unclear risk

This studies used PCA morphine in addition to the study drugs

Pagnoni 1996 Methods

Randomised double-blind trial, three arms.

Participants

62 women above the age of 18 years who were ASA I or II undergoing a caesarean section were recruited into the study Contraindications were women with history of bleeding disorders, chronic drug abuse, contraindications to aspirin or ibuprofen use

Interventions

Group 1 (30 women): oral ibuprofen arginine 400 mg. Group 2 (30 women): intramuscular ketorolac 30 mg -this group was not included as the review is on oral analgesia Group 3 (32 women): placebo. Rescue medication was intramuscular ketoprofen 100 mg.

Outcomes

Pain intensity using VAS scores, side effects.

Notes Oral analgesia for relieving post-caesarean pain (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

30

Pagnoni 1996

(Continued)

Risk of bias Bias




Not described except “randomly assigned numbers”.


Low risk

“the clinical investigator was provided with sealed envelopes specifying the treatment given; these to be opened only in an emergency.”


The study medication as well as a rescue pain relief were administered by clinical investigator


Clinical investigator administered the drugs and adverse effects were recorded by a physician


Low risk

Data complete.


Unclear risk


Other bias

Unclear risk

1 study used ketoprofen 100 mg intramuscularly as rescue medication Adverse events (none) reported only for the 6-hour study period

Palmer 2000 Methods

Randomised double-blind trial, two arms.

Participants

60 ASA I and II women undergoing non-urgent caesarean delivery under spinal anaesthesia

Interventions

Oral clonidine 0.1 mg every 12 hours and control group received placebo at same intervals All participants received IV morphine via PCA pump for 24 hours post-operatively for 24 hours There were 60 women enrolled, but no data on how many women in each group

Outcomes

PCA morphine use, side effects such as hypotension, nausea and pruritus

Notes

Only 1 patient from 1 of the groups required treatment for hypotension Results for additional pain relief were added as dosages for PCA in milligrams, data from this study were not be used in the meta-analysis


31

Palmer 2000

(Continued)

Risk of bias Bias




No description of randomisation.


Unclear risk

Not described.


Not described.


Not described.


High risk

The data presented in such form that it is not clear if there were indeed any missing data as the total numbers are not provided


Unclear risk


Other bias

High risk

This study used PCA morphine in addition to the study drugs. It was only published in abstract form in 2000

Sammour 2011 Methods

Randomised trial, four arms.

Participants

120 women who delivered at Bnai-Zion Medical Center between August 7th 2006 and March 23rd 2009 who had either an elective or emergency caesarean section

Interventions

Tramadol 100 mg every 6 hours orally, oral tramadol 100 mg on request, oral naproxen 500 mg every 8 hours and oral naproxen 500 mg on request Post-operatively parenteral morphine was given to all participants. Oral paracetamolpropoxyphene was given to those requiring additional pain relief 30 women in each group.

Outcomes

Pain using VAS scores, side effects, mobility, breastfeeding and need for rescue medication

Notes

Data could not be used as the definition for adequate pain relief used by the authors is different to the definition in this review

Risk of bias Bias




32

Sammour 2011

(Continued)


Computer-generated randomisation numbers were used.


Sequentially numbered opaque envelopes with the codes sealed inside were used

Low risk

Blinding of participants and personnel High risk (performance bias) All outcomes

Open labelled.


Open labelled.


Low risk

10 women dropped out from tramadol at fixed intervals group and five dropped out from tramadol on request group, but the data were presented inclusive of these women


Unclear risk


Other bias

Unclear risk

This study used parenteral morphine in the first two hours post-operatively and the ward on request they received oral paracetamol-dextropoxyphene

Short 2012 Methods

Randomised placebo-controlled double-blind trial. three arms

Participants

132 ASA 1 or 2 women with singleton pregnancy at term admitted for elective caesarean delivery were enrolled after informed consent, six women were excluded after randomisation (2 protocol breach, 3 failed spinal, 1 postponed) Exclusion criteria were contraindications to neuraxial anaesthesia or to any of the medication used in the study, history of epilepsy, central nervous system or mental disorder, chronic pain, drug abuse, or use of neuropathic analgesics or antiepileptic drugs

Interventions

Gabapentin 600 mg, Gabapentin 300 mg and placebo. 42 women in each group

Outcomes

Pain intensity at rest and on movement using VAS scores, nausea, vomiting, sedation, pruritus, difficulties with breastfeeding and incisional pain at 3 months





33

Short 2012

(Continued)


Computer-generated randomisation tables in blocks of 6.


Low risk

Sequentially numbered envelopes containing 300 mg Gabapentin or lactose placebo in an identical blue cover were used


Hospital pharmacist placed the study medications in identical capsules


Study personnel administering medications was blinded. Personnel performing post-operative assessments were not blinded to the results


Low risk

2 women were excluded from each group after receiving intervention due to failed spinal (1 in each group), protocol breach (1 in each intervention group, and 1 due to protocol breach in control group


Low risk

Trial protocol registered at www.clinicaltrials.gov NCT01094925

Other bias

Unclear risk

This study used 2 mg of IV morphine at 5 minutes interval for the first 2 hours in PACU, then 50 mg oral diclofenac every 8 hours and 1 g acetaminophen every 6 hours for 72 hours. Breakthrough pain was managed with 2 mg subcutaneous morphine on request and subsequently received 10 mg oral morphine as required Underpowered study.

Sunshine 1992 Methods

Randomised trial, four arms

Participants

161 post-caesarean women of legal age who were able to communicate meaningfully with a nurse-observer and who gave consent were considered for the study

Interventions

Tramadol 75 mg (40 women), tramadol 150 mg (40 women), acetaminophene 650 mg and propoxyphene (40 women) and placebo (41 women)

Outcomes

Pain relief and adverse effects.


34

Sunshine 1992

(Continued)

Notes

Data for adequacy of pain relief could not be used, standard deviation for the means (SPID) not stated. This study contributed data on need for additional analgesia and adverse reactions for main comparison and subgroup analysis (tramadol 50 versus 100 mg)

Risk of bias Bias




Computer-generated randomisation in blocks of 8. Parallel design. 2 women received each treatment


Low risk

Following randomisation each patient received a single unit dose, consisting of 2 blue opaque capsules


Not described - but all unit doses had identical appearance and packaging. Stated that ”medication was administered in “double-blind manner”


Not described.


Low risk

No missing data reported.


Unclear risk


Other bias

Unclear risk

Use of rescue medication.

Sunshine 1993 Methods

Randomised trial, five arms.

Participants

240 women who were at least 18 years old and who reported severe post-operative pain after caesarean section Exclusion criteria included, hypersensitivity to any of the study drugs, breastfeeding mothers, serious physical or mental illness, history of drug or alcohol abuse or those who received any other investigational drug within 1 month before enrolment in the study

Interventions

Group 1: 100 mg ketoprofen. Group 2: 50 mg ketoprofen. Group 3: 650 mg acetaminophen plus 10 mg oxycodone hydrochloride Group 4: 650 mg acetaminophen. Group 5: placebo. 48 women in each group.

Outcomes

Pain intensity and side effects.


35

Sunshine 1993

(Continued)

Notes

Data not presented in full, e.g. no SD for means. Complex study design, e.g. 5 groups. women requiring re medication before 1 hour after the first dose of study medication were dropped from the study and replaced. 3 treatment groups in the repeated-dose phase

Risk of bias Bias




Methods of sequence generation were not described. Parallel design. Women who received acetaminophen alone or placebo were re-assigned randomly to receive 1 of the 2 doses of ketoprofen or acetaminophen with oxycondone


All drugs were identical in appearance.

Low risk


Personel were blinded.


Assessors were blinded to the group allocation.


High risk

3 women did not enter repeated doses study; 13 did not receive repeat medications for 2 days and were excluded, 10 participants withdrew from repeated doses study because they did not have pain, 2 withdrew due to fever and 1 withdrew due to reasons unrelated to the study


Unclear risk


Other bias

High risk

The design of the trial with 5 groups and second randomisation is complex and likely to have introduced bias. Participants received repeated doses if they complained of pain 2 hours after receiving study medication

COX-2: cyclooxygenase 2 inhibitor IV: intravenous NICU: neonatal intensive care unit NSAIDs: non-steroidal anti-inflammatories PACU: postanaesthesia care unit PCA: patient-controlled analgesia SD: standard deviation SPID: summed pain intensity difference VAS: visual analogue score VPS: verbal pain scores Oral analgesia for relieving post-caesarean pain (Review) Copyright © 2015 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

36

Characteristics of excluded studies [ordered by study ID]

Study

Reason for exclusion

Abboud 1990

Prophylactic oral naltrexone was compared with intrathecal morphine for caesarean section. This review assesses oral analgesia only, and does not include intrathecal analgesia

Alhashemi 2006

The study assessed intravenous acetaminophen versus oral ibuprofen and in our review only assess oral analgesia for post-caesarean pain

Bloomfield 1993

The study assessed women undergoing caesarean section or abdominal hysterectomy and the results were not interpreted separately for each group. Our review only assesses oral analgesia for post-caesarean pain

Carvalho 2007

This study assessed Extended-Release Epidural Morphine compared with conventional morphine for post-caesarean pain and our review only assesses oral analgesia for post-caesarean pain

Carvalho 2008

This is only a protocol of the study, but not the study itself

Choi 2003

The study assessed oral dextrometorphan and intrathecal morphine for analgesia after caesarean section and our review only assesses oral analgesia for post-caesarean pain

Danesh 2009

This study assessed oral methadone and intramuscular pethidine for post-caesarean pain and our review only assesses oral analgesia

Davis 2006

This study assessed oral analgesics compared to intravenous PCA for post-caesarean pain and our review assesses oral analgesia

De Lia 1986

This study assessed women post-caesarean section, post vaginal and abdominal hysterectomy and the results were not interpreted separately. Our review only assesses oral analgesia for post-caesarean pain

Dieterich 2012

The study assesses oral analgesics and intravenous patient-control analgesia and in our review we are only looking at oral analgesics

East 2007

This study assesses standard administered analgesia and self-administered but in our review we do not have this subgroup

Esper 2005

This study assesses Valdecoxib for reducing post-operative opioid use and patient-control analgesia morphine is used for pain relief but our review assesses oral analgesics for post-caesarean pain.

Ganem 2003

The study assesses effects of low spinal morphine associated to intravenous and oral ketoprofen post-caesarean section and in our review we are looking at only oral analgesia for post-caesarean pain

Gharabaghi 2011

This study assesses natural plant Rosa damascene for pain relief and in our review only assesses pharmacological drugs

Liu 2011

Combined oral oxycodone hydrochloride controlled-release tablets plus paracetamol and tramadol was compared with epidural analgesia with respect to uterine cramping, pain control and side effects after caesarean section.


37

(Continued)

Epidural analgesia is not part of our review Marzida 2009

The study assessed oral diclofenac and subcutaneous pethidine but our review only assesses oral analgesia

McDonnell 2010

The study assessed oral oxycodone and intrathecal morphine but our review only assesses oral analgesia

Moore 2011

In this study 28% of women were excluded because of protocol violation

Morrison 1986

This study assessed women post-caesarean section, post vaginal and abdominal hysterectomy but the results were not interpreted separately

Nakou 2010

This study compared oral celecoxib and patient-controlled epidural analgesia, which is not part of our review

Olson 1984

The study assessed women post episiotomy, dysmenorrhoea and post-caesarean section and the results were not interpreted for each group

Ortner 2011

This study has been published as a letter to the publisher, the study was terminated after enrolling 5 women because of a high VAS score

Paech 2014

This study assessed the effectiveness of adjunct analgesia with patient-controlled epidural analgesia following a caesarean delivery. The interventions included a combination of intravenous and oral medication versus placebo and, therefore, the results could not be interpreted separately for oral analgesia

Palangio 2000

This study assessed a combination of hydrocodone and ibuprofen versus oxycodone and acetaminophen for post-operative obstetrics and gynaecological pain including caesarean section but the results were not interpreted separately

Samimi 2011

The study assessed celecoxib for prevention of morphine-induced pruritus in women undergoing caesarean section and in our review we are looking at oral analgesics for post-caesarean pain

Shah 2010

This study assessed effectiveness of maternal gabapentin therapy at the time of delivery of neonatal pain response during intramuscular injection, which is not part of our review

Sibilla 1994

This study looked at different types of morphine administration for post-caesarean pain and in our review we are looking at only oral analgesics for post-caesarean pain

Snell 2006

This study looked at oral morphine patient or midwife-administered and intramuscular morphine for postcaesarean pain and in our review we are looking at oral analgesia for post-caesarean pain

Stuart 1995

Epidural opioids were compared with an oral regimen to reduce pruritis after epidural morphine. Epidural is not part of our review

Sunshine 1983

This study assessed ketoprofen or aspirin or placebo for post partum pain, i.e. post episiotomy or post-caesarean section or postpartum uterine cramping and our review assesses post-caesarean pain

Sunshine 1985a

This study assessed oral ibuprofen with and without codeine and placebo for incisional pain post episiotomy or post-caesarean section and our review assesses post-caesarean pain


38

(Continued)

Sunshine 1985b

This study assessed oral analgesia for postpartum pain and results for women who had a caesarean section were not interpreted separately

Sunshine 1986

The study assessed oral Ketoprofen, aspirin and placebo for postpartum pain including caesarean section but results are not interpreted separately

Sunshine 1987

The study assessed pain resulting from episiotomy, caesarean section, and gynaecological surgery and the results were not presented separately for each group

Sunshine 1988

The study looked oral Pentazocine versus pentazocine-Naloxone for post-operative pain and results were not interpreted separately for caesarean section

Sunshine 1989

In this study women with episiotomy, post-caesarean and uterine cramp were studied, however, the data for postcaesarean pain were not presented separately

Sunshine 1997a

Hydrocodone and ibuprofen were compared for post-caesarean, post-gynaecological surgery and the data were not interpreted separately

Sunshine 1997b

The study assessed hydrocodone with ibuprofen and ibuprofen alone for acute post-operative pain including caesarean section but results are not interpreted separately for caesarean section and in our review we are looking at oral analgesics for post-caesarean pain

Tan 2007

The study assessed oral celecoxib or intravenous ketamine or their combination for post-caesarean pain and in our review we only assess oral analgesia

Wittels 1993

This study looked at effects of epidural butorphanol, epidural nalbuphine and oral naltrexone on post-caesarean epidural morphine and post-surgery women were randomised to intravenous PCA with either morphine or meperidine. In our review we are only looking at oral analgesia for post-caesarean pain

Yanagidate 2001

This study assessed clonidine premedication, to see if it reduces morphine requirements after a caesarean section without affecting newborn’s outcome and our review assesses oral analgesia for post-caesarean pain

PCA: patient-controlled analgesia VAS: visual analogue scale


39

DATA AND ANALYSES

Comparison 1. Opioid versus placebo

Outcome or subgroup title 1 Need for additional pain relief with a different drug 2 Adverse drug effects

No. of studies

No. of participants

1

120

Risk Ratio (M-H, Fixed, 95% CI)

0.33 [0.06, 1.92]

1

120


6.58 [0.38, 113.96]

Statistical method

Effect size

Comparison 2. Non-opioid analgesics versus placebo

Outcome or subgroup title 1 Need for additional pain relief 1.1 Celexocib versus placebo 1.2 Gabapentin versus placebo 1.3 Ibuprofen versus placebo 1.4 Ketoprofen versus placebo 1.5 Naproxen versus placebo 1.6 Paracetamol versus placebo 2 Maternal drug effects

No. of studies

No. of participants

6 1 1 1 1 1 2

584 60 126 62 120 80 136

Risk Ratio (M-H, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI)

0.70 [0.48, 1.01] 0.89 [0.59, 1.35] 0.34 [0.23, 0.51] 0.66 [0.41, 1.07] 1.05 [0.86, 1.28] 0.11 [0.01, 2.00] 0.77 [0.43, 1.40]

2

267


11.12 [2.13, 58.22]

Statistical method

Effect size

Comparison 3. Combination versus placebo

Outcome or subgroup title 1 Need for additional pain relief 1.1 Paracetamol plus codeine versus placebo 1.2 Paracetamol plus oxycodone versus placebo 1.3 Paracetamol plus propoxyphene versus placebo 2 Maternal drug effects

No. of studies

No. of participants

3 1

242 65

Risk Ratio (M-H, Random, 95% CI) Risk Ratio (M-H, Random, 95% CI)

0.70 [0.35, 1.40] 0.44 [0.23, 0.82]

1

96

Risk Ratio (M-H, Random, 95% CI)

1.0 [0.78, 1.28]

1

81

Risk Ratio (M-H, Random, 95% CI)

0.65 [0.11, 3.69]

3

252


13.18 [2.86, 60.68]

Statistical method


Effect size

40

Comparison 4. Opioid versus non-opioid

Outcome or subgroup title 1 Need for additional pain relief with a different drug 2 Maternal adverse effects

No. of studies

No. of participants

1

121


0.51 [0.07, 3.51]

2

241


2.32 [1.15, 4.69]

Statistical method

Effect size

Comparison 5. Opioid analgesics versus combination analgesics

Outcome or subgroup title 1 Need for additional pain relief 1.1 Tramadol versus paracetamol plus propoxyphene 2 Maternal adverse effects

No. of studies

No. of participants

1 1

121 121

Risk Ratio (M-H, Fixed, 95% CI) Risk Ratio (M-H, Fixed, 95% CI)

0.51 [0.07, 3.51] 0.51 [0.07, 3.51]

1

121


6.74 [0.39, 116.79]

Statistical method

Effect size

Comparison 6. Non-opioid versus combination analgesics

Outcome or subgroup title 1 Need for additional pain relief with a different drug

No. of studies

No. of participants

1

192

Statistical method Risk Ratio (M-H, Fixed, 95% CI)

Effect size 0.87 [0.81, 0.93]

Comparison 7. Non-opioid analgesics versus placebo (subgroup analysis by high and low doses of the same drug)

Outcome or subgroup title 1 Need for additional pain relief 1.1 Gapapentin 300 mg 1.2 Gabapentin 600 mg 1.3 Ketoprofen 50 mg 1.4 Ketoprofen 100 mg

No. of studies

No. of participants

2 1 1 1 1

63 63 72 72

Statistical method Risk Ratio (M-H, Fixed, 95% CI) Risk Ratio (M-H, Fixed, 95% CI) Risk Ratio (M-H, Fixed, 95% CI) Risk Ratio (M-H, Fixed, 95% CI) Risk Ratio (M-H, Fixed, 95% CI)


Effect size Subtotals only 0.25 [0.13, 0.49] 0.44 [0.27, 0.71] 0.83 [0.64, 1.07] 0.55 [0.39, 0.79]

41

Comparison 8. Opioid analgesics versus placebo (subgroup analysis by high and low doses of the same drug)

Outcome or subgroup title 1 Need for additional pain relief 1.1 Tramadol 75 mg 1.2 Tramadol 150 mg

No. of studies

No. of participants

1 1 1

80 80

Statistical method

Effect size

Risk Ratio (M-H, Fixed, 95% CI) Risk Ratio (M-H, Fixed, 95% CI) Risk Ratio (M-H, Fixed, 95% CI)

Subtotals only 0.67 [0.12, 3.78] 0.14 [0.01, 2.68]

Analysis 1.1. Comparison 1 Opioid versus placebo, Outcome 1 Need for additional pain relief with a different drug. Review:

Oral analgesia for relieving post-caesarean pain

Comparison: 1 Opioid versus placebo Outcome: 1 Need for additional pain relief with a different drug

Study or subgroup

Opioid

Placebo

n/N

n/N

Risk Ratio

Weight

Sunshine 1992

2/80

3/40

100.0 %

0.33 [ 0.06, 1.92 ]

Total (95% CI)

80

40

100.0 %

0.33 [ 0.06, 1.92 ]

M-H,Fixed,95% CI

Risk Ratio M-H,Fixed,95% CI

Total events: 2 (Opioid), 3 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 1.23 (P = 0.22) Test for subgroup differences: Not applicable

0.01

0.1

Favours opioid

1

10

100

Favours placebo


42

Analysis 1.2. Comparison 1 Opioid versus placebo, Outcome 2 Adverse drug effects. Review:


Comparison: 1 Opioid versus placebo Outcome: 2 Adverse drug effects

Study or subgroup

Opioid

Placebo

n/N

n/N

Risk Ratio

Weight

Sunshine 1992

6/80

0/40

100.0 %

6.58 [ 0.38, 113.96 ]

Total (95% CI)

80

40

100.0 %

6.58 [ 0.38, 113.96 ]

M-H,Fixed,95% CI

Risk Ratio M-H,Fixed,95% CI

Total events: 6 (Opioid), 0 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 1.29 (P = 0.20) Test for subgroup differences: Not applicable

0.01

0.1

Favours opioid

1

10

100

Favours placebo


43

Analysis 2.1. Comparison 2 Non-opioid analgesics versus placebo, Outcome 1 Need for additional pain relief. Review:


Comparison: 2 Non-opioid analgesics versus placebo Outcome: 1 Need for additional pain relief

Study or subgroup

Non-opioids

Placebo

Risk Ratio MH,Random,95% CI

Weight


n/N

n/N

17/30

19/30

15.9 %

0.89 [ 0.59, 1.35 ]

30

30

15.9 %

0.89 [ 0.59, 1.35 ]

22/84

32/42

16.2 %

0.34 [ 0.23, 0.51 ]

84

42

16.2 %

0.34 [ 0.23, 0.51 ]

13/30

21/32

14.9 %

0.66 [ 0.41, 1.07 ]

30

32

14.9 %

0.66 [ 0.41, 1.07 ]

84/96

20/24

19.0 %

1.05 [ 0.86, 1.28 ]

96

24

19.0 %

1.05 [ 0.86, 1.28 ]

0/40

4/40

1.5 %

0.11 [ 0.01, 2.00 ]

40

40

1.5 %

0.11 [ 0.01, 2.00 ]

14/43

12/21

13.5 %

0.57 [ 0.32, 1.01 ]

1 Celexocib versus placebo Lee 2004

Subtotal (95% CI) Total events: 17 (Non-opioids), 19 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 0.53 (P = 0.60) 2 Gabapentin versus placebo Short 2012

Subtotal (95% CI) Total events: 22 (Non-opioids), 32 (Placebo) Heterogeneity: not applicable

Test for overall effect: Z = 5.27 (P < 0.00001) 3 Ibuprofen versus placebo Pagnoni 1996

Subtotal (95% CI) Total events: 13 (Non-opioids), 21 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 1.69 (P = 0.090) 4 Ketoprofen versus placebo Sunshine 1993

Subtotal (95% CI) Total events: 84 (Non-opioids), 20 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 0.49 (P = 0.62) 5 Naproxen versus placebo Angle 2002

Subtotal (95% CI) Total events: 0 (Non-opioids), 4 (Placebo) Heterogeneity: not applicable Test for overall effect: Z = 1.49 (P = 0.14) 6 Paracetamol versus placebo Bjune 1996

0.01

0.1

Favours non-opioid

1

10

100

Favours control

(Continued . . . )


44

(. . . Study or subgroup

Non-opioids

Sunshine 1993

Placebo


Weight

Continued)


n/N

n/N

40/48

21/24

18.9 %

0.95 [ 0.78, 1.16 ]

91

45

32.5 %

0.77 [ 0.43, 1.40 ]

100.0 %

0.70 [ 0.48, 1.01 ]

Subtotal (95% CI) Total events: 54 (Non-opioids), 33 (Placebo)

Heterogeneity: Tau2 = 0.14; Chi2 = 3.99, df = 1 (P = 0.05); I2 =75% Test for overall effect: Z = 0.85 (P = 0.40)

Total (95% CI)

371

213

Total events: 190 (Non-opioids), 129 (Placebo) Heterogeneity: Tau2 = 0.18; Chi2 = 40.77, df = 6 (P

Postcaesarean section analgesia: are opioids still required?

EMLA(R) cream: a pain-relieving strategy for childhood vaccination.

Relieving the pain of pressure sores.

Early analgesia for acute abdominal pain.

Relieving pain using dose-extending placebos: a scoping review.

Cancer pain in America: missing the target of relieving suffering.

The pain-relieving qualities of exercise in knee osteoarthritis.

Postcaesarean open-heart surgery for Streptococcus sanguinis infective endocarditis.

Oral ketamine analgesia in chronic pain and problematic rise in blood pressure.

Effect of caffeine on ibuprofen analgesia in postoperative oral surgery pain.

Patient satisfaction between remifentanil patient-controlled analgesia and epidural analgesia for labor pain.

Oral gabapentin versus rectal diclofenac for postoperative analgesia.

Sudden persistent fetal bradycardia after spinal analgesia for labor pain.

Analgesia for tonic pain by self-administered lateral hypothalamic stimulation.

Analgesia for formalin-induced pain by lateral hypothalamic stimulation.

Neurobiological Mechanism of Acupuncture for Relieving Visceral Pain of Gastrointestinal Origin.

The role for epigenetic modifications in pain and analgesia response.

Patient-controlled spinal analgesia for postoperative pain control.

Back massage intervention for relieving lower back pain in puerperal women: A randomized control trial study.

Preventive analgesia for postoperative pain control: a broader concept.

The efficacy of cold-gel packing for relieving episiotomy pain - a quasi-randomised control trial.

Intraperitoneal dexamethasone as a new method for relieving postoperative shoulder pain after gynecologic laparoscopy.

What is the most effective treatment for relieving the pain of a jellyfish sting?

Relieving Pain in St. Joseph, Missouri: A Health Services Partnership for Changing the Way Pain is Perceived, Judged and Treated.