Trauma is often accompanied by pain—whether that is major trauma or a single injury (Association of Ambulance Chief Executives (AACE), 2016). Providing adequate analgesia to reduce this is a simple humanitarian aim (AACE, 2016). There have been considerable developments for this subset of patients over the past 25 years, such as the introduction of major trauma networks; however, there are still improvements that can be made (National Institute for Health and Care Excellence (NICE), 2016). One key clinical area highlighted by NICE (2016) is pain management in patients with traumatic injuries. NICE (2016) notes that those suffering pain should have immediate and effective pain relief and, without it, patients can experience delayed healing, chronic pain and disability.
Trauma in context
Major trauma is a minor element of the total workload in emergency care, with estimations of less than 0.2% of the total workload (National Audit Office (NAO), 2010). Trauma such as dislocation, fracture, joint injury, and amputation were the second most common diagnoses in accident and emergency between 2015 and 2016 (Baker, 2017). The true costs of trauma are unknown when taking into account immediate and subsequent treatment, support, and loss of economic output (NAO, 2010; NICE, 2016). There is room for development in pain management as pain relief is often inadequately treated in traumatic injuries leading to oligoanalgesia, the undertreatment of pain (Parker and Rodgers, 2015).
Prehospital analgesia
Ambulance services in the UK have been using morphine sulphate for analgesia since 2004 (Joint Royal Colleges Ambulance Liaison Committee (JRCALC), 2004). Morphine's introduction into the ambulance setting has faced several barriers including legislative and legal hurdles, such as the possession and administration of opioids (Lord and Nicholls, 2014). Compared with Australia, where morphine was introduced in the 1980s, its establishment in the UK is comparatively late (Lord and Nicholls, 2014). Opiates, especially morphine, have been considered the standard analgesia in patients experiencing severe pain (Lord and Nicholls, 2014), with NICE (2016) recommending intravenous morphine as the first-line analgesic in patients with traumatic injuries. There is no consensus or evidence-based data for the optimal choice or dose of opiate analgesia in the prehospital environment (Dijkstra et al, 2014).
The recent development of specialist paramedic roles and changes in the law relating to controlled drugs has meant that the analgesic options available to paramedics have progressed, and will continue to evolve (Hodkinson, 2016). There are many elements to pain; for example:
This literature review will aim to discover what is currently known, and identify any gaps and areas for future research concerning opioid pain relief in adults with traumatic injuries out-of-hospital.
Methodology
To formulate the clinical focus for the current literature review, a Population Intervention Comparison Outcome (PICO) framework was used (Table 1). Initially, the Journal of Paramedic Practice (JPP) online database was searched using the broad term ‘pain’. Grey literature was also examined from UK government and relevant UK health organisations. The Cochrane Library was searched and had only one protocol that related specifically to this review (Metcalfe et al, 2015); however, this was retracted in March 2017 owing to the authors not intending to complete the review (Wiley Online Library, 2017). The following databases were searched through the University of Brighton's online library between February and March 2017:
| Population | Intervention | Comparison | Outcome |
|---|---|---|---|
| Conscious adults with traumatic pain. | Opioid analgesia. | Of intervention variables. | Reduced pain levels, |
The search terms used are shown in Table 2, though multiple searches were conducted including or excluding different combinations. The initial searches were limited to the past 5 years, 2012–2017. However, because of the lack of out-of-hospital research, this was extended to the past 10 years (2007–2017). Table 3 shows the results from these searches.
| Search 1 | Search 2 | Search 3 |
|---|---|---|
| Pre hospital or prehospital or pre-hospital or out of hospital or out-of-hospital or emergency department or emergency setting | Pre hospital or prehospital or pre-hospital or out of hospital or out-of-hospital or emergency department or emergency setting | Pre hospital or prehospital or pre-hospital or out of hospital or out-of-hospital or emergency department or emergency setting |
| AND | AND | AND |
| Trauma* or polytrauma* or poly trauma* or multiple trauma* or accident* or injur* or fall* | Intravenous or IV | Intranasal or IN |
| AND | AND | AND |
| Pain management or pain relief | Morphine | Morphine |
| AND | AND | AND |
| Morphine or fentanyl or opi* | Fentanyl | Fentanyl |
| NOT | NOT | NOT |
| Child* or pediatric* or paediatric* | Child* or pediatric* or paediatric* | Child* or pediatric* or paediatric* |
| CINAHL | MEDLINE | EMBASE | SCOPUS | |
|---|---|---|---|---|
| Search 1 | 247 |
111 | 262 | 1097 |
| Search 2 | 41 | 46 | 91 | 70 |
| Search 3 | 6 | 63 | 204 | 312 |
The searches were also filtered manually using a variety of criteria to determine relevance (Grove et al, 2015). Inclusion and exclusion criteria are shown in Table 4. A manual search of the reference lists of the literature provided additional resources (Aveyard et al, 2015). Both qualitative and quantitative studies were found; however, the quantitative studies will form the main body of the current research owing to the comparative line of enquiry. To analyse the research, the adapted ‘six questions for critical thinking’ framework was used as it was designed for health care and has been adapted from various feedback of the tool (Aveyard et al, 2015). See Table 5 for study comparisons.
| Inclusion | Exclusion |
|---|---|
| Studies between 2012–2017 |
Studies 10> years |
| Study type | Sample size | Findings | |
|---|---|---|---|
| Smith et al (2012) | Non-randomised double blinded clinical trial | 200 |
No significant difference in analgesic effectiveness between morphine and fentanyl. |
| Fleischman et al (2010) | Retrospective before and after study | 718 |
Morphine and fentanyl produced similar out-of-hospital analgesia, though fentanyl needed higher doses. |
| Bounes et al (2010) | Randomised double blind clinical trial | 108 |
Sufentanil had a slight earlier onset, but is not superior to morphine for traumatic pain relief in out-of-hospital environment. Adequate dosing is the key to pain relief. |
| Rickard et al (2007) | Randomised open label clinical trial | 258 |
No significant difference in the effectiveness of intranasal fentanyl and intravenous morphine. |
| Middleton et al (2010) | Retrospective cohort comparative study | 42 844 |
Both intravenous morphine and intranasal fentanyl were effective analgesics. |
Morphine vs. fentanyl
Different opioid, different effects
Opioids have been shown in many studies to be safe and effective in treating pain in the emergency setting (Dijkstra et al, 2014). It is unknown which opioid is the most effective, and to what extent they individually lead to adverse events (Niemi-Murola et al, 2011). There are multiple options of opioid, such as diamorphine and hydromorphine (Patanwala et al, 2010). Each opioid has its own distinctive properties; for example, morphine is hydrophilic compared to fentanyl, which is lipophilic (Niemi-Murola et al, 2011). These differences have different effects on the body, such as onset time and duration of analgesia (Thomas and Shewakramani, 2008). The out-of-hospital literature located concerned only morphine and fentanyl (Fleischman et al, 2010; Smith et al, 2012), with one study comparing morphine to sufentanil—a derivative of fentanyl (Bounes et al, 2010).
Morphine is the opioid of choice for patients in acute pain, including those with traumatic injuries, and is often compared against in studies (Lord and Nicholls, 2014). Bounes et al (2008) compared two different morphine doses out-of-hospital and found that they were both effective at relieving pain at 30 minutes; however, the higher dose was more beneficial at 10 minutes, with minimal side-effects. There was no statistical significant difference in adverse effects between both the groups, although the higher-dose group did have a higher incidence of adverse effects.
Fentanyl, a synthetic opioid, is meant to be more potent and have a faster onset of analgesia, though it is shorter-acting when compared with morphine (Niemi-Murola et al, 2011). Owing to its rapid onset and offset, it is considered beneficial in the prehospital setting (Friesgaard et al, 2016). Administration of fentanyl by non-physician ambulance personnel has also been shown to be effective and safe (Friesgaard et al, 2016).
Looking at opioids comparatively: part 1
Though Bounes et al (2008), Soriya et al (2012) and Friesgaard et al (2016) looked at their respective opioids individually in effect and safety in the out-of-hospital setting, the drugs were not compared against each other to find the optimal evidence-based opioid.
Currently, the most recent out-of-hospital clinical trial of morphine vs. fentanyl is the Smith et al (2012) study. The main findings were that there was no significant difference in analgesic effect or occurrence of adverse effects between intravenous (IV) morphine and IV fentanyl. These results are echoed in Bounes et al (2010) and Fleischman et al (2010).
With the Smith et al (2012) study, although measurement of numeric pain score (NPS) and administration of medication was supposed to happen every 5 minutes, this was presented as not being fulfilled with no detailed breakdown or reasoning. This suggests there was a possible measurement error, author bias, or unknown confounder such as other patient interventions taking priority (Grove et al, 2015). Despite the fact that there is no perfect rigour in research, there are aspects that could have been taken into account with the dependant variable to improve reliability and validity (Grove et al, 2015).
Bounes et al (2010) were able to show more rigour in their study, as they were able to measure and present the findings at 3-minute intervals until 12 minutes; and then at 15 and 30 minutes after enrolment onto the study. Data were incomplete with only 6 patients from Bounes et al (2010); however, they provided reasons compared to Smith et al (2012) who had unaccounted inconsistencies.
The sample size used by Smith et al (2012) was appropriately powered (n=80) (Polgar and Thomas, 2013; Grove et al, 2015) and they exceeded their desired number of participants (Smith et al, 2012). However, compared with a previous study by Bounes et al (2010), Smith et al (2012) had a larger sample size; with any non-significant findings, detailed attention should be paid to the sample size and power retrospectively as there can be a risk of a type-two error; meaning they would be unable to detect any significant difference rather than finding no significant difference (Grove et al, 2015). Smith et al (2012) did not carry out any further analysis of their sample or powering; however, it is mentioned that enrolment of patients was limited and that the authors enrolled most of the patients themselves. This suggests potential of a larger sample that might have improved the effect size, as well as the research being at further risk of author bias (Grove et al, 2015).
Looking at opioids comparatively: part 2
Fleischman et al (2010) conducted a retrospective analysis before and after the study. Although considered lower in the hierarchy of evidence (National Health and Medical Research Council (NHMRC), 2009), different situations affect what evidence is appropriate and, as with a systematic literature review, a mixed-method approach might be preferred (Aveyard et al, 2015). The context for Fleischman et al (2010) was a protocol change within an ambulance service. The benefits this had over Smith et al (2012) and Bounes et al (2010) was that it was multicentre, which would increase its representativeness of the target population (Grove et al, 2015). This also assisted in obtaining a larger sample size—Fleischman et al (2010) attained 718 participants compared with Smith et al (2012) with 200, and Bounes et al (2010) with 108. This meant they had a better effect size and reduced observer bias (Grove et al, 2015). However, with non-experimental research, there are challenges, such as reduced rigour, which can manifest as missing data. This was the case with Fleischman et al (2010); approximately 30% of data within the two data sets was missing, confounding the results.
Smith et al (2012) had a directional hypothesis but the results showed no statistical significance between the two study drugs, confirming the null hypothesis (Grove et al, 2015). A confounding factor mentioned by Smith et al (2012) in their results was that 57.5% of their patients had received analgesics before enrolment onto the study. This contamination, especially as there was no documentation of previous dosing, makes the potential of a type-two error more probable.
Intranasal administration
The out-of-hospital application of intranasal (IN) analgesia is a recent development within the UK (Hodkinson, 2016). Although it has been used in other countries, there is still limited evidence available for its use (Hansen and Dahl, 2013).
The IN route is highly vascularised with a permeable membrane that can bypass the blood-brain barrier straight to the central nervous system (CNS) via the olfactory nerve (Lötsch et al, 2013). The potential advantages over the IV route are the rapid onset of analgesia, assisted by direct CNS delivery; and the ease of administration (Prommer and Thompson, 2011).
The out-of-hospital IN research concentrates on fentanyl (Karlsen et al, 2014). Although morphine has been studied previously (DeNatale et al, 2010), drug absorption through the nasal mucosal is dependent on lipophilicity (Prommer and Thompson, 2011). Fentanyl's lipophilic properties allow it to quickly pass between plasma and the CNS (Hansen and Dahl, 2013). Fentanyl is reportedly 100 times more potent than morphine with IN formulations—this means a smaller volume is needed to achieve the same analgesic effect. This reduces the risk of ‘run-off’ after dispensing the analgesia, meaning a more accurate dose is given (Lötsch et al, 2013).
Rickard et al (2007) is currently the highest level of evidence available for a quantitative comparative study within this topic area (Aveyard et al, 2015; Grove et al, 2015). The Rickard et al (2007) non-blinded multicentre trial showed that there was no significant difference between IN fentanyl or IV morphine in effectiveness of analgesia from baseline to destination.
The design of this research is presented with a detailed intervention strategy and comprehensive data collection, aiding in the rigour and validity of this study (Polgar and Thomas, 2013; Grove et al, 2015). One negating factor in the design was that the group was heterogeneous in terms of patients' sources of pain (Grove et al, 2015). This meant that no one patient cohort could be analysed specifically, although orthopaedic injuries were the most common complaint (Rickard et al, 2007).
The main countering factor with the Rickard et al (2007) study is the sample size. To have a power of 80, 200 patients per group were needed to detect any difference according to Rickard et al (2007). A total of 227 patients were evaluated compared to the desired 400. Similarly to Smith et al (2012), Rickard et al (2007) are at high risk of a type-two error, where they were unable to detect any significant difference (rather than finding no significant difference) (Grove et al, 2015).
Another addition to strengthen the possibility of a type-two error is the contamination of results with a variety of additional analgesics used in the study. This is accepted as a limitation by Rickard et al (2007), who are more accountable than Smith et al (2012) and Middleton et al (2010) with similar problems; however, this confounds the results nonetheless.
This unresolved comparison of administration route is continued with later research, such as the Middleton et al (2010) retrospective study that is the most up-to-date study comparing IV morphine to IN fentanyl. Middleton et al (2010) conclude that IV morphine appears more effective; however, that is with multivariate logistic regression analysis. When looking at mean effectiveness between the two opioids and administration routes, there were no significant differences. There were also many confounding factors. For example, no data on timing or doses given were analysed; only a subset of paramedics could administer the fentanyl and 12 955 patients were analysed in the morphine group compared with 3778 for the fentanyl group.
Discussion
Where the research is now
The heterogeneity between the current small numbers of studies inhibits any potential meta-analysis (Niemi-Murola et al, 2011; Dijkstra et al, 2014; Grove et al, 2015). One area with variation is the dosing and strategy of titration of the fentanyl and morphine, which differs between studies as is shown in Table 6. Even within some of the studies themselves, there was a variety of doses given; there was no rationale given for the protocol used; or the data were not presented (Rickard et al, 2007; Bounes et al, 2010; Fleischman et al, 2010; Middleton et al, 2010; Smith et al, 2012).
| Study | IV Morphine (Repeat Dose) | IV Morphine Max | IV Fentanyl (Repeat dose) | IV Fentanyl Max | IN Fentanyl (Repeat dose) | IN Fentanyl Max |
|---|---|---|---|---|---|---|
| Smith et al (2012) | 4 mg |
20 mg | 50 μg |
250 μg | ||
| Fleischman et al (2010) | 2.5–5 mg |
20 mg | 50 mcg |
200 μg | ||
| Bounes et al (2010) | 0.15 mg/kg |
Until pain relieved | (Sufentanil) 0.15 mcg/kg |
Until pain relieved | ||
| Rickard et al (2007) | 2.5–5 mg |
15 mg | 180 μg |
420 μg | ||
| Middleton et al (2010) | 5 mg |
0.5 mg/kg | 240 μg |
No maximum |
mg = milligrams, μg = micrograms, kg = kilograms,
A common impression throughout the work reviewed is the need for more research in the field of out-of-hospital pain management (Friesgaard et al, 2016). The author of the current paper considers this to be a potential reason why the Metcalfe et al (2015) protocol for The Cochrane Library was halted in March 2017 (Wiley Online Library, 2017) owing to a lack of research for a systematic review.
Alongside effectiveness and safety, the onset and duration of analgesia would benefit from an increased comparative focus, as the studies so far have not found any significant difference between opioid or administration route.
Other considerations
Another element for consideration is the longstanding financial pressure the NHS faces (Dunn et al, 2016). NICE (2016) provides the cost analysis as shown in Table 7. In terms of the way NICE (2016) evaluates prices in the context of their recommendations, if there is no clinical difference between the opioids, cost might lean their preference towards fentanyl.
| Item | Cost |
|---|---|
| Intranasal equipment | £3.19 |
| Intravenous equipment | £2.00 |
| 10 milligram morphine ampoule | £0.94 |
| 100 microgram fentanyl ampoule | £0.30 |
Currently, in the UK, only physicians can deliver fentanyl (NICE, 2016). The administration of fentanyl by paramedics and other ambulance personnel has been shown to be safe elsewhere however (Friesgaard et al, 2016). The development of specialist paramedics and recent changes in UK law for the use of controlled drugs has enabled paramedics to have a broader amount of analgesic options (Hodkinson, 2016). Historically, barriers in law have been eased, as in the case of the introduction of morphine into practice (Lord and Nicholls, 2014). Potential for changes to the law regarding fentanyl are not unrealistic. Collen (2016) highlights that independent prescribing by advanced paramedics would require changes to legislation.
The Commission on Human Medicines was unable to accept independent prescribing owing to inconsistencies with role titles and the need for improved definition of advance practice nationally (Collen, 2016). One potential avenue for this review focus could be through patient group directions (PGDs). Collen (2016) notes that PGDs are suitable in critical care, but less useful with patients that are elderly and have comorbidities. PGDs could enable the use of fentanyl by paramedics, but Collen's (2016) concerns with more complex patient conditions crosses over into critical care, particularly considering the rise of major trauma in older patients (The Trauma Audit & Research Network (TARN), 2017).
Protocols in place
Lord and Nicholls (2014) note that increasing the range of analgesic options alone is unlikely to lead to significant improvements. Their conclusions are reinforced by further studies that reveal how the protocols of analgesia were not fully used (Friesgaard et al, 2016). Oligoanalgesia is apparent in patients with traumatic pain out-of-hospital with either unadministered or administered analgesia. Situational factors, as well as characteristics of both patients and medical staff, create variation in treatment (Albrecht et al, 2013). French et al (2013) conclude that paramedic knowledge, perceptions, and management improve with continual education on pain. Education of paramedics alongside organisational improvements, such as protocols in pain management, could improve oligoanalgesia (Lord and Nicholls, 2014).
Fentanyl is suggested to have a faster onset (Friesgaard et al, 2016) and is cheaper than morphine (NICE, 2016), making it ideal in the out-of-hospital environment. However, it should be noted that the onset time of fentanyl has not been proven to be better than morphine through any out-of-hospital clinical trials. Bounes et al (2010) did find that sufentanil had a faster onset time than morphine. Fentanyl has also been shown to have been safely used by paramedics (Friesgaard et al, 2016). There are some instances where IN would not be appropriate, such as in facial trauma, but there are also potential weaknesses with limited absorption and inadequate duration (Fleischman et al, 2010). This would suggest that initial IN administration before attempting IV access would be sensible if possible. Smith et al (2012) suggest that either IV fentanyl or morphine could be used. If fentanyl has the benefit of quicker onset and lower cost, this may precede morphine in preference—especially if IN administration was not established first.
Morphine is currently still the recommended first-line treatment (AACE, 2016; NICE, 2016) so any changes would have to show significant improvements to justify any alteration in UK law and ambulance protocols.
Whether fentanyl or morphine is used, it is documented that individuals have different opioid tolerances and a protocol of titration should account for those differences (Bounes et al, 2010). Once an ideal formulation of opioid and administration route is found, a focus on delivering the analgesia to patients would aim to reduce oligoanalgesia. This could be through changes in pain-management protocols alongside continued education for paramedics.
Limitations
The current literature review used a systematic approach. As a result of the original research being carried out as part of the author's undergraduate project, a comprehensive review was beyond its capabilities. Additionally, this meant that a narrow focus was maintained, even though there are many elements to pain such as alternative and combinations of analgesics (AACE, 2016). There was also a paucity of strong evidence available (Aveyard et al, 2015), alongside the heterogeneity between the small number of studies, inhibiting any potential meta-analysis (Dijkstra et al, 2014; Grove et al, 2015).
Conclusion
Treating pain is important for humanitarian reasons, as it may prevent patient deterioration; allow for better assessment; and limit potential post injury complications (NICE, 2016). Although major trauma is encountered infrequently, minor trauma is common. Pain is also one of the most common symptoms to present to the UK ambulance services (AACE, 2016).
The studies explored show that morphine IV and both fentanyl IV and IN significantly reduce pain effectively, but that there is no significant difference between them. It was also found that fentanyl could be safely used by ambulance personnel (Friesgaard et al, 2016).
There was a lack of up-to-date out-of-hospital research available. Of the suitable studies, the majority were inferior to RCTs in terms of the accepted hierarchy of evidence, and those that were RCTs had confounding elements (Aveyard et al, 2015). Although there is no perfect research, steps could be taken to improve what has already been done (Grove et al, 2015). Any new studies would benefit from a change in focus from effectiveness to onset, duration, and dose of analgesia. This would allow the question of what is the most appropriate out-of-hospital opioid in adults with traumatic pain to be answered more coherently.
For any change regarding paramedic use of fentanyl, amendment to UK laws would be required. The current review shows that more research is needed before any changes are considered. Out-of-hospital pain management is far from ideal with oligoanalgesia prevalent within some of the studies (Bakkelund et al, 2013). Continual education of paramedics and quality improvement from an organisational perspective in pain management could improve current practice, as well as patient outcomes and experiences (Lord and Nicholls, 2014).
Recommendations
In terms of future directions for research, further out-of-hospital RCT research comparing IV morphine to both IV and IN fentanyl is recommended, as well as a change of research focus from effectiveness to onset, duration and dose of analgesia. Research into continued educational programmes in pain management for paramedics is also needed.