Skip to main content
NCBI home page
The Cochrane Database of Systematic Reviews logoLink to The Cochrane Database of Systematic Reviews
. 2017 Mar 2;2017(3):CD011863. doi: 10.1002/14651858.CD011863.pub2

Pre‐hospital opioid analgesia for traumatic injuries

David Metcalfe 1,✉, Olubode A Olufajo 2, Ali Salim 2
PMCID: PMC6464689

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To assess the benefits and harms of different opioid pharmacological agents given in the pre‐hospital setting for pain caused by traumatic injury.

Background

Description of the condition

Pain is commonly experienced following traumatic injury. However, there is a range of pharmacological options for treating pain before the patient reaches hospital. These include paracetamol, non‐steroidal anti‐inflammatory drugs (e.g. ibuprofen, indomethacin, diclofenac), opioids, nitrous oxide, and ketamine (Dijkstra 2014). Such drugs may be administered through a number of different routes (e.g. oral, rectal, intravenous, intramuscular) but vary in terms of their availability, efficacy, and side effect profiles.

Despite these therapeutic options, traumatic pain is under‐treated, particularly in the pre‐hospital setting (Luger 2003; Jennings 2011). Barriers to adequate pre‐hospital analgesia include failure to report pain (Gunnarsdottir 2002), patient refusal of pharmacological treatment (Singer 2008; Iqbal 2013), and knowledge deficits among care providers (Tanabe 2000). Health care professionals may be particularly reluctant to administer opioids in the pre‐hospital setting for fear of adverse events. For example, one qualitative study found that paramedics preferred not to use fentanyl (or administered only ineffective doses) and emergency nurses were less likely to give intravenous opioids in the absence of a physician (Berben 2012). A British study found that less than a quarter of patients transported to a hospital by ambulance with fractures received pre‐hospital opioid analgesia (Siriwardena 2010).

Uncontrolled pain is unsatisfactory from a humanitarian perspective (IASP 2011) but can also impair immune function, healing, and functional recovery (Sacerdote 2000). Inadequate analgesia in the acute setting may also contribute to the development of chronic pain (Wang 2009). Up to 63% of patients report chronic pain one year after suffering major trauma (Rivara 2008).

Description of the intervention

Opioids have been used in the treatment of pain for thousands of years (Trescot 2008). They are a class of chemicals that work through opioid receptors found principally in the central nervous system, peripheral nervous system, and gastrointestinal tract. Their natural agonists are the endogenous opioid peptides (e.g. endorphins, enkephalins, and dynorphins) but a range of agents that act on these receptors are now available.

The term “opiate” properly refers to the naturally occurring alkaloids extracted from the resin of the opium poppy (Papaver somniferum). However, “opioid” is a broader term that includes opiates, synthetic substances that function through opioid receptors, and opioid peptides such as the endorphins (Laux‐Biehlmann 2013).

Opioids have been categorized into a number of broad classes that include:

  • Natural opiates: alkaloids from the resin of the opium poppy that include morphine, codeine, and thebaine.

  • Morphine esters: opioids derived from morphine that include diacetylmorphine (diamorphine or “heroin”).

  • Semi‐synthetic opioids: chemically derived from other natural opiates or from morphine esters. These include oxycodone and buprenorphine.

  • Fully synthetic opioids: chemically synthesized drugs designed to stimulate opioid receptors. These include fentanyl, pethidine, and methadone.

  • Atypical opioids: those not easily classified, such as tramadol, that act on opioid receptors but also through a number of non‐opioid pathways.

  • Opioid peptides: naturally produced peptides within the human body (e.g. endorphins) that have a poorly understood role in determining mood and motivation.

Opioids may also be classified in other ways, e.g. by their specific opioid receptors, mechanism of action (agonist, partial agonist, antagonist, partial antagonist), or by schedule under specific legal jurisdictions.

How the intervention might work

Opioid receptors are distributed throughout the central and peripheral nervous systems as well as peripherally, e.g. around the gastrointestinal tract. These receptors are classified into three major subtypes: μ (mu‐opioid receptor, MOR), κ (kappa‐opioid receptor, KOR), and δ (delta‐opioid receptor, DOR) (Trescot 2008; Allouche 2014).

Most commonly used opioid analgesic drugs (including morphine) are opioid agonists, i.e. they stimulate opioid receptors (Allouche 2014). However, a small number (e.g. buprenorphine) are partial agonists that stimulate only a partial response on receptor binding. Specific opioids vary in terms of their relative stimulation of opioid receptors, bioavailability, and metabolism (Trescot 2008). Although opioids are known to act directly on the central nervous system, it is increasingly recognized that their analgesic effects might be partially modulated by stimulation of peripheral opioid receptors (Sehgal 2011).

In addition to their analgesic properties, opioids are associated with a range of physiological effects. Frequently occurring side effects include nausea, vomiting, constipation, urinary retention, reduced consciousness, and respiratory depression. These may be a particular concern in the pre‐hospital trauma setting. For example, vomiting poses an aspiration risk to patients whose cervical spine is immobilized and reduced consciousness may confuse the clinical picture in the context of head injury. There is also evidence from animal models that morphine exerts a number of cardiovascular properties (affecting both heart rate and blood pressure) during active haemorrhage (Kirkman 2014). Morphine also increases mortality during prolonged haemorrhagic hypovolaemic shock in rats (Molina 2004).

Why it is important to do this review

There is continued uncertainty about which pharmacological agents provide optimal pain relief for injured patients. In particular, it is unclear which opioid analgesics relieve pain most effectively in the pre‐hospital trauma setting and have the most favourable safety profiles. This uncertainty may contribute to the continued under‐treatment of traumatic pain and obstruct the development of pre‐hospital protocols and guidelines.

A systematic review is therefore necessary both to guide clinical practice and to identify directions for further research.

Objectives

To assess the benefits and harms of different opioid pharmacological agents given in the pre‐hospital setting for pain caused by traumatic injury.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCTs) designed to assess the effectiveness of opioids in injured patients before reaching hospital. RCTs will include quasi‐randomised trials, which use a method of allocating participants that is not strictly random, e.g. alternate days. We will only include trials that were prospectively registered unless the final report was published before 2010.

Types of participants

All patients (regardless of age) that have suffered any form of injury, i.e. bodily damage caused by physical trauma. No thresholds for injury severity (e.g. Injury Severity Score) will be employed.

Injury will be interpreted broadly (e.g. including burns) and no specific injured groups (e.g. isolated hip fractures) will be excluded. However, "injury" diagnoses not caused by acute trauma (e.g. repetitive strain injury) will be excluded. Studies with mixed populations (e.g. all patients transported to hospital) will be excluded unless the proportion of non‐injured patients is small and comparable between groups, or the data for injured patients are presented separately.

Types of interventions

Opioids describe a broad class of drugs that may be categorised in different ways. We will include any opioid given for the relief of pain, at any dose, and given by any route of administration.

The main comparisons are opioid analgesia versus other non‐opioid analgesic agents. However, a number of sub‐group analyses are also planned (Subgroup analysis and investigation of heterogeneity).

It is likely that opioids will be administered in combination with other pain‐relieving interventions. We will include studies using combination analgesia if adjunct pharmacological therapies are distributed equally (e.g. by protocol) between the intervention groups. It is possible that additional drugs will be administered by health care professionals because pain was not relieved by the original intervention. If this routinely occurs prior to assessment of the primary outcome, we will include these studies but consider such cases as evidence of treatment failure (see Subgroup analysis and investigation of heterogeneity).

Concurrent conservative interventions (e.g. splinting a fractured limb) will be acceptable as it is assumed these will be used in all groups across all studies.

Types of outcome measures

We will include all studies that satisfy the Criteria for considering studies for this review, even if they do not report our pre‐selected outcome measures. All primary and secondary outcome measures will be reported in a 'Summary of findings table' once the review is completed.

Primary outcomes

The primary outcome will be the proportion of patients reporting "minimal pain", which has been defined in the literature as a Visual Analogue Scale (VAS) score of <30/100mm (Moore 2013; Wiffen 2013). VAS results may be presented using different scales (e.g. 1‐5, 1‐100) and so all data will be scaled to 100 for comparison.

VAS is often recorded at a series of time points following drug administration. The primary outcome measure will use the VAS score recorded at 10 minutes post‐administration. This is based on research that showed the mean time to effect of intravenous morphine is 5 minutes (Tveita 2008) and that the majority (89%) of senior doctors believe that pain relief should be achieved within 10 minutes (Park 2010).

If pain is not recorded at 10 minutes, the closest available time point will be accepted. In the event of there being two time periods equidistant to 10 minutes, e.g. 5 and 15 minutes, the later measure will be used.

If 50% of patients receiving an active drug achieve minimal pain (VAS <30/100) compared with 40% in the control group, we would require a total sample size of 1,032 patients to detect this pooled intervention effect with 90% power at the 5% significance level.

Secondary outcomes

  • Change in VAS (from baseline) at 10, 20, 30, 60, and 120 minutes.

  • Adverse events (e.g. vomiting) attributed by study authors to the intervention.

  • Need for intubation or opioid reversal attributed to respiratory depression by the study authors.

  • Mortality (pre‐ or in‐hospital).

  • Patient‐reported quality of life (e.g. EQ‐5D, SF‐12).

  • Glasgow Outcome Score (GOS) or extended GOS (GOS‐E).

Search methods for identification of studies

In order to reduce publication and retrieval bias we will not restrict our search by language, date or publication status.

Electronic searches

The Cochrane Injuries Group Trials Search Co‐ordinator will search the following:

  1. Cochrane Injuries Group specialised register (present version);

  2. The Cochrane Library (http://www.cochranelibrary.com/) (latest issue);

  3. Ovid MEDLINE(R), Ovid MEDLINE(R) In‐Process & Other Non‐Indexed Citations, Ovid MEDLINE(R) Daily and Ovid OLDMEDLINE(R) (1946 to present);

  4. Embase Classic + Embase (OvidSP) (1947 to present);

  5. PubMed (http://www.ncbi.nlm.nih.gov/pubmed/) (present);

  6. ISI Web of Science: Science Citation Index Expanded (SCI‐EXPANDED) (1970 to present);

  7. ISI Web of Science: Conference Proceedings Citation Index‐Science (CPCI‐S) (1990 to present);

  8. LILACS (http://lilacs.bvsalud.org/) (present);

  9. PROSPERO (http://www.crd.york.ac.uk/prospero/search.asp) (present);

  10. Clinicaltrials.gov (www.clinicaltrials.gov);

  11. WHO International Clinical Trials Registry Platform (http://apps.who.int/trialsearch/).

We will adapt the MEDLINE search strategy (Appendix 1) as necessary for each of the other databases: the added study filter is a modified version of the Ovid MEDLINE Cochrane Highly Sensitive Search Strategy for identifying randomised trials and to the Embase search strategy we will add the study design terms as used by the UK Cochrane Centre (Lefebvre 2011).

Searching other resources

Reference lists from full text articles will be handsearched for further studies not identified by the electronic searches.

Data collection and analysis

Selection of studies

Two review authors (DM and OO) will independently screen the abstracts of retrieved search items for potentially eligible studies. Full text articles will then be downloaded and assessed for inclusion by these authors using the inclusion and exclusion criteria above. Disagreements will be resolved by discussion with a third author (AS) available to arbitrate if uncertainty persists.

Data extraction and management

A data extraction form will be created and piloted. Two review authors (DM and OO) will use this form to extract data from the included studies. Disagreements will be resolved through discussion with recourse to a third author (AS) if necessary.

Assessment of risk of bias in included studies

Two authors (DM and OO) will independently assess each included study using the Cochrane 'risk of bias' tool (Higgins 2011). This tool includes selection bias (sequence generation, allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessment), attrition bias (incomplete outcome data), and reporting bias (selective outcome reporting).

Measures of treatment effect

Continuous data (e.g. VAS scores) will be expressed as mean differences (MD) with 95% confidence intervals (CI). Dichotomous outcomes (e.g. adverse events) will be expressed as relative risk ratios with 95% CI. We will also aim to present treatment effects as "number needed to treat".

Unit of analysis issues

No specific unit of analysis issues are anticipated. The unit of analysis for this review will be the individual patient.

Dealing with missing data

Trial investigators will be contacted to request any missing data. The primary analysis will only consider those data points that are available, i.e. excluding the missing data. However, there is a concern that some studies may have substantial loss to follow‐up, which will be explored using sensitivity analyses. This is described further in Sensitivity analysis.

Similarly, if trial authors report having imputed outcome measures (e.g. last observation carried forward), we will attempt to conduct sensitivity analyses to explore the effect of using these techniques. If imputed outcomes are reported frequently by included studies, we will report this finding in our Results and Discussion sections. If standard deviations (SDs) are not specifically reported, we will determine these from standard errors, confidence intervals, or exact P values where possible. We do not plan to impute standard deviations if there are insufficient data to calculate them precisely.

Assessment of heterogeneity

Heterogeneity between studies will be assessed using a visual inspection of confidence intervals on forest plots. We will also calculate a Chi2 statistic (with P < 0.05 suggestive of significant heterogeneity) and an I2 statistic. The I2 statistic will be interpreted as: 0‐30% possibly unimportant, 31‐60% moderate heterogeneity, 61‐90% substantial heterogeneity, and 91‐100% very substantial heterogeneity.

Assessment of reporting biases

A funnel plot will be created and visually inspected for asymmetry if at least 10 studies are available for meta‐analysis. The usefulness of funnel plots for identifying publication bias is diminished when the number of available studies is small (Higgins 2011).

Data synthesis

Previous systematic reviews have identified substantial heterogeneity between studies (Park 2010) and so we plan to pool data from trials using the random‐effects model. In the event of very substantial heterogeneity (I2 > 75%), we will consider forgoing the meta‐analysis in favour of a narrative synthesis.

Subgroup analysis and investigation of heterogeneity

Comparisons selected a priori for subgroup analysis if there are sufficient data include:

  • Patients with isolated injuries (e.g. single fracture or laceration) versus patients with multiple injuries (i.e. injuries to more than one anatomical system)

  • Children (age <18) versus adults (age 18 and older)

  • Opioid receptor agonists versus partial agonists

  • Different opioids as categorised above in Description of the intervention, i.e. natural opiates, morphine esters, semi‐synthetic opioids, fully synthetic opioids, and atypical opioids.

If sufficient data are available, we will compare individual opioid drugs given at different doses, times, and routes of administration.

Sensitivity analysis

Sensitivity analyses will be used to investigate specific aspects of trial methodology. In particular, sensitivity analyses will be performed to investigate the impact of including studies that are at high risk of selection bias, i.e. quasi‐randomised trials and trials with inadequate allocation concealment. In the event of trials inadequately blinding patients and/or health care professionals, the effect of including these will also be explored using sensitivity analyses.

Sensitivity analyses will also be used to explore the effects of missing data (Dealing with missing data). In the first sensitivity analysis, missing dichotomous outcomes will be treated as failures and missing continuous data points as lying at the extreme of the distribution (2 SDs). A more moderate position will be adopted in a second sensitivity analysis in which missing dichotomous outcomes will be excluded and missing continuous data treated as lying at 1 SD from the mean.

Trial sequential analysis methods will be used to explore the result of any statistically significant effect found for the review's primary outcome that did not meet the expected sample size (Roberts 2015).

Acknowledgements

This project was supported by the UK National Institute for Health Research, through Cochrane Infrastructure funding to the Cochrane Injuries Group. The views and opinions expressed are those of the authors and do not necessarily reflect those of the Systematic Reviews Programme, NIHR, NHS or the Department of Health.

Appendices

Appendix 1. Search strategy

Ovid MEDLINE(R), Ovid MEDLINE(R) In‐Process & Other Non‐Indexed Citations, Ovid MEDLINE(R) Daily and Ovid OLDMEDLINE(R) (1946 to present) 1. exp "Wounds and Injuries"/ 2. (injuries or injury).ab,ti. 3. injuries.fs. 4. 1 or 2 or 3 5. Analgesics, Opioid/ 6. (opioid* or opiate*).ti,ab. 7. 5 or 6 8. 4 and 7 9. randomi?ed.ab,ti. 10. randomized controlled trial.pt. 11. controlled clinical trial.pt. 12. placebo.ab. 13. clinical trials as topic.sh. 14. randomly.ab. 15. trial.ti. 16. Comparative Study/ 17. 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 18. (animals not (humans and animals)).sh. 19. 17 not 18

What's new

Date Event Description
2 March 2017 Amended The authors do not intend to complete this review. The protocol is being withdrawn from the Cochrane Library.
Open in a new tab

Contributions of authors

DM drafted the protocol.

OO critically revised the protocol for content and performed the information size calculation.

AS guided key methodological decisions and critically revised the protocol for content.

Sources of support

Internal sources

  • No sources of support supplied

External sources

  • Royal College of Surgeons of England Fulbright Scholarship, UK.

    Salary and expenses support for David Metcalfe

Declarations of interest

DM: None known.

OO: None known.

AS: None known.

Notes

2 March 2017: The authors do not intend to complete this review. The protocol is being withdrawn from the Cochrane Library.

Withdrawn from publication for reasons stated in the review

References

Additional references

  1. Allouche S, Noble F, Marie N. Opioid receptor desensitization: mechanisms and its link to tolerance. Frontiers in Pharmacology 2014;5:280. [PUBMED: 25566076] [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Berben SA, Meijs TH, Grunsven PM, Schoonhoven L, Achterberg T. Facilitators and barriers in pain management for trauma patients in the chain of emergency care. Injury 2012;43(9):1397‐2. [PUBMED: 21371708] [DOI] [PubMed] [Google Scholar]
  3. Dijkstra BM, Berben SA, Dongen RT, Schoonhoven L. Review on pharmacological pain management in trauma patients in (pre‐hospital) emergency medicine in the Netherlands. European Journal of Pain 2014;18(1):3‐19. [PUBMED: 23737462] [DOI] [PubMed] [Google Scholar]
  4. Gunnarsdottir S, Donovan HS, Serlin RC, Voge C, Ward S. Patient‐related barriers to pain management: the Barriers Questionnaire II (BQ‐II). Pain 2002;99(3):385‐96. [PUBMED: 12406513] [DOI] [PubMed] [Google Scholar]
  5. Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. [Google Scholar]
  6. International Pain Summit of the International Association for the Study of Pain. Declaration of Montréal: declaration that access to pain management is a fundamental human right. Journal of Pain & Palliative Care Pharmacotherapy 2011;25(1):29‐31. [PUBMED: 21426215] [DOI] [PubMed] [Google Scholar]
  7. Iqbal M, Spaight PA, Siriwardena AN. Patients' and emergency clinicians' perceptions of improving pre‐hospital pain management: a qualitative study. Emergency Medicine Journal 2013;30(3):e18. [PUBMED: 22544875] [DOI] [PubMed] [Google Scholar]
  8. Jennings PA, Cameron P, Bernard S. Epidemiology of prehospital pain: an opportunity for improvement. Emergency Medicine Journal 2011;28(6):530‐1. [PUBMED: 20679429] [DOI] [PubMed] [Google Scholar]
  9. Kirkman E, Watts S. Haemodynamic changes in trauma. British Journal of Anaesthesia 2014;113(2):266‐75. [PUBMED: 25038158] [DOI] [PubMed] [Google Scholar]
  10. Laux‐Biehlmann A, Mouheiche J, Vérièpe J, Goumon Y. Endogenous morphine and its metabolites in mammals: history, synthesis, localization and perspectives. Neuroscience 2013;233:95‐117. [PUBMED: 23266549] [DOI] [PubMed] [Google Scholar]
  11. Lefebvre C, Manheimer E, Glanville J. Chapter 6: Searching for studies. In: 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.
  12. Luger TJ, Lederer W, Gassner M, Löckinger A, Ulmer H, Lorenz IH. Acute pain is underassessed in out‐of‐hospital emergencies. Academic Emergency Medicine 2003;10(6):627‐32. [PUBMED: 12782523] [DOI] [PubMed] [Google Scholar]
  13. Molina PE, Zambell KL, Zhang P, Vande Stouwe C, Carnal J. Hemodynamic and immune consequences of opiate analgesia after trauma/hemorrhage. Shock 2004;21(6):526‐34. [PUBMED: 15167681] [DOI] [PubMed] [Google Scholar]
  14. Moore RA, Straube S, Aldington D. Pain measure and cut‐offs ‐ 'no worse than mild pain' as a simple, universal outcome. Anaesthesia 2013;68(4):400‐12. [PUBMED: 23347230] [DOI] [PubMed] [Google Scholar]
  15. Park CL, Roberts DE, Aldington DJ, Moore RA. Prehospital analgesia: systematic review of evidence. Journal of the Royal Army Medical Corps 2010;156(4 Suppl 1):295‐300. [PUBMED: 21302646] [DOI] [PubMed] [Google Scholar]
  16. Rivara FP, Mackenzie EJ, Jurkovich GJ, Nathens AB, Wang J, Scharfstein DO. Prevalence of pain in patients 1 year after major trauma. Archives of Surgery 2008;143(3):282‐7. [PUBMED: 18347276] [DOI] [PubMed] [Google Scholar]
  17. Roberts I, Ker K, Edwards P, Beecher D, Manno D, Sydenham E. The knowledge system underpinning health care is not fit for purpose and must change. BMJ 2015;350:h2463. [DOI] [PubMed]
  18. Sacerdote P, Bianchi M, Gaspani L, Manfredi B, Maucione A, Terno G, et al. The effects of tramadol and morphine on immune responses and pain after surgery in cancer patients. Anesthesia & Analgesia 2000;90(6):1411‐4. [PUBMED: 10825330] [DOI] [PubMed] [Google Scholar]
  19. Sehgal N, Smith HS, Manchikanti L. Peripherally acting opioids and clinical implications for pain control. Pain Physician 2011;14(3):249‐58. [PUBMED: 21587328] [PubMed] [Google Scholar]
  20. Singer AJ, Garra G, Chohan JK, Dalmedo C, Thode HC Jr. Triage pain scores and the desire for and use of analgesics. Annals of Emergency Medicine 2008;52(6):689‐95. [PUBMED: 18501475] [DOI] [PubMed] [Google Scholar]
  21. Siriwardena AN, Shaw D, Bouliotis G. Exploratory cross‐sectional study of factors associated with pre‐hospital management of pain. Journal of Evaluation in Clinical Practice 2010;16(6):1269‐75. [PUBMED: 20722889] [DOI] [PubMed] [Google Scholar]
  22. Tanabe P, Buschmann M. Emergency nurses' knowledge of pain management principles. Journal of Emergency Nursing 2000;26(4):299‐305. [PUBMED: 10940837] [DOI] [PubMed] [Google Scholar]
  23. Trescot AM, Datta S, Lee M, Hansen H. Opioid pharmacology. Pain Physician 2008;11:S133‐53. [PUBMED: 18443637] [PubMed] [Google Scholar]
  24. Tveita T, Thoner J, Klepstad P, Dale O, Jystad A, Borchgrevink PC. A controlled comparison between single doses of intravenous and intramuscular morphine with respect to analgesic effects and patient safety. Acta Anaesthesiologica Scandinavica 2008;52(7):920‐5. [PUBMED: 18702754] [DOI] [PubMed] [Google Scholar]
  25. Wang CK, Myunghae Hah J, Carroll I. Factors contributing to pain chronicity. Current Pain and Headache Reports 2009;13(1):7‐11. [PUBMED: 19126364] [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Wiffen PJ, Wee B, Moore RA. Oral morphine for cancer pain. Cochrane Database of Systematic Reviews 2013, Issue 7. [DOI: 10.1002/14651858.CD003868.pub3; PUBMED: 23881654] [DOI] [PubMed] [Google Scholar]

Articles from The Cochrane Database of Systematic Reviews are provided here courtesy of Wiley

RESOURCES